Trip Distribution: Theoretical Background

Chapter 14
Trip Distribution
In t h is ch a pt er , th e m ech a n ics of t h e secon d crim e t r a vel dem a n d m odeling st a ge -
t r ip d ist r ibu t ion - is expla in ed. T rip distribu tion is a m odel of th e num ber of tr ips th at
occu r bet ween ea ch or igin zon e a n d ea ch des t in a t ion zon e. It u ses th e p red ict ed nu m ber of
t r ip s or igin a t in g in ea ch or igin zon e (t r ip pr odu ct ion m odel) a n d t h e pr edict ed n u m ber of
t r ips en din g in ea ch d est in a t ion zon e (tr ip a t t r a ction m odel). Th u s, t r ip d ist r ibu t ion is a
m odel of t r a vel bet ween zon es - tr ips or link s. The m odeled t r ip dist r ibut ion ca n t h en be
com p a r ed t o t h e a ct u a l d is t r ibu t ion t o s ee wh et h er t h e m od el p r od u ces a r ea s on a ble
a p pr oxim a t ion .
Th e o r e t i c a l B a c k g ro u n d
Th e t h eor et ical ba ckgr oun d beh in d t h e t r ip d ist r ibu t ion m odu le is pr es en t ed firs t .
Next , t h e specific pr ocedu r es a n d t est s a r e discu ss ed wit h t h e m odel bein g illus t r a t ed wit h
da t a from Ba lt im ore Coun t y.
Lo g ic o f t h e Mo d e l
Tr ip dis t ribu t ion u su a lly occu r s t h r ou gh a n a lloca t ion m od el t h a t sp lit s t r ip s fr om

each origin zone into distinct destina tions. Tha t is, th ere is a m at rix which r elat es the
n u m ber of t r ips origin a t in g in ea ch zon e t o th e n u m ber of t r ips en din g in ea ch zon e. F igu r e
14.1 illust r a t es a t ypica l ar r a n gemen t . In t h is m a t r ix, th er e a r e a n u m ber of or igin zon es,
M, an d a n u m ber of dest ina t ion zon es, N. The origin zon es in clud e a l l t h e dest in a t ion
zon es bu t m a y also in clud e some a ddit ion a l on es. The r ea sons t h a t t h er e wou ld be
d iffer en t n u m ber s of zon es for t h e or igin a n d d es t in a t ion m od els a r e t h a t cr im e d a t a for
oth er jur isd ictions a r e n ot a va ila ble bu t t h a t a sizea ble n u m ber of crim es t h a t occur r ed in
t h e st u dy ju r isd iction a r e com m it t ed by in dividu a ls wh o lived t h ose oth er jur isd ictions. If it
wer e possible t o obt a in crim e da t a for t h e City of Balt imore, t h en it would be pr efer a ble to
h a ve th e sa m e n u m ber of zon es for bot h t h e or igin file an d t h e dest ina t ion file.
F or exam ple, wit h t h e Balt imore Coun t y dat a t h a t a r e being us ed t o illus t r a t e t h e

m odel, th er e a r e 325 dest ina t ion s zon es for Balt imore Coun t y while th e or igin zon es
in clu de bot h t h e 325 in Ba lt im or e Cou n t y a n d 207 m or e fr om t h e a dja cen t Cit y of
Ba lt im ore. As cha pt er 12 poin t ed out , t h e st u dy a r ea sh ould ext en d be yon d t h e m odelin g
a r ea u n t il t h e origin s of at lea st 95% of a ll t r ips en din g in t h e st u dy a r ea a r e cou n t ed.
E a ch cell in t h e m a t r ix in dica t es t h e n u m ber of trips t h a t go fr om ea ch or igin zon e

t o ea ch dest in a t ion zon e. To u se t h e exa m ple in figu r e 14.1, t h er e wer e 15 t r ip s fr om zon e
1 t o zone 2, 21 t r ips from zon e 1 t o zone 3, an d s o fort h . Not e t h a t t h e t r ips a r e
a sym m et r ical; th a t is, t r ips in one d ir ection a r e differ en t t h a n t r ips in t h e opposit e
dir ect ion . To u se t h e t a ble, t h er e wer e 15 t r ip s fr om zon e 1 t o zon e 2, bu t on ly 7 t r ip s fr om
zon e 2 to zon e 1.
14.1
Th e t r ips on t h e dia gon a l ar e intra-zonal t r ips , t r ips t h a t origin a t e a n d en d in t h e
sam e zone. Again, to use th e exam ple below, th ere were 37 tr ips th at both originat ed and
en ded in zon e 1, 53 t r ip s t h a t bot h or igin a t ed a n d en ded in zon e 2, a n d so for t h .
In su ch a m odel, cons t a n cy is m a in t a in ed in t h a t t h e n u m ber of t r ips origin a t in g

fr om a ll or igin s zon es m u st equ a l t h e n u m ber of t r ip s en d in g in a ll d es t in a t ion zon es . Th is
is t h e fun da m en t a l ba la n cing equ a t ion for a t r ip d ist r ibu t ion . In equ a t ion for m , it is
expressed as:
M N
G Oi = G Dj (14.1)
I=1 j=1
wh er e t h e origin s, O i , a r e su m m ed over M origin zones w h ile t h e des t in a t ions , D j, a r e

su m m ed over N dest in a t ion zon es. To u se t h e exa m ple below, t h e t ota l n u m ber of origin s
is equa l to th e t ot a l nu m ber of dest ina t ion s, an d is equ a l to 43,240
F igu r e 14.1
Exam ple Crime Origin-De stin ation Matrix
Th e ba la n cing equ a t ion is im plem en t ed in a ser ies of st eps t h a t in clud e m odelin g

t h e n u m ber of cr imes or igina t ing in ea ch zon e, add ing in t r ips origin a t ing fr om ou t side t h e
st u dy a r ea (ext er n a l t r ip s), a n d st a t is t ica lly ba la n cin g t h e or igin s a n d dest in a t ion s so t h a t
14.2
equ a t ion 14.1 h olds . Th is wa s d one in t h e t r ip gen er a t ion st a ge. Bu t , it is es sen t ia l t h a t
t h e st ep sh ou ld h a ve been com plet ed for t h e t r ip dist r ibut ion t o be implem en t ed.
Ob se rv e d a n d P r e di ct e d D is tri bu ti on s
Th er e a r e t wo tr ip d ist r ibu t ion m a t r ices t h a t n eed t o be dist in gu ish ed. Th e firs t is

t h e observed (or em p ir ica l) d is t ribu t ion . Th is is th e a ct u a l n u m ber of t r ip s t h a t a r e
obser ved t r a velin g bet ween ea ch origin zon e a n d ea ch d est in a t ion zon e. In gen er a l, wit h
cr im e d at a , s uch a n em p ir ica l d is t ribu t ion wou ld be obt a in ed fr om a n a r r es t r ecor d wh er e
t h e r esiden ce (or a r r est ) loca t ion of ea ch offen der is list ed for ea ch cr ime t h a t t h e offen der
wa s cha r ged with . In t h is ca se, th e r esiden ce/a r r est loca t ion wou ld be con sider ed t h e
origin wh ile t h e crim e loca t ion would be cons ide r ed t h e des t in a t ion.
In ch a pt er 12, it wa s m en t ion ed t h a t t h er e is alwa ys un cer t a int y as t o t h e t r u e

or igin loca t ion of a cr im e in cid en t , wh et h er t h e offen der a ct u a lly t r a veled fr om t h e
r es ide n ce locat ion t o th e crim e loca t ion or even wh et h er t h e offen der wa s a ctu a lly livin g a t
t h e r esid en ce loca t ion . Bu t a bsen t a n y a lt er n a t ive eviden ce, a m ea n in gfu l d is t r ibu t ion ca n
st ill be obta in ed by sim ply t r ea t in g t h e r esiden ce locat ion a s a n a pp r oxim a t e origin .
Th e obser ved d ist r ibu t ion is calcula t ed by sim ply en u m er a t in g t h e n u m ber of t r ips
by each origin -dest ina t ion com bina t ion . This is somet imes ca lled a t rip lin k (or t r ip pa ir ).
Th er e a r e n ot a n y special st a t ist ics oth er t h a n a sim ple t wo-wa y cr oss-clas sifica t ion t a ble.
Th e secon d d ist r ibu t ion , however , is a m odel of t h e t r ip d ist r ibu t ion m a t r ix. This is
u su a lly ca lled th e predicted distr ibut ion. In th is case, a simple model is used to
a ppr oxim a t e t h e a ct u a l emp irical dist r ibut ion . The t r ips origin a t ing in ea ch or igin zon e a r e
a llocat ed t o dest in a t ion zon es u su a lly on t h e ba sis of bein g dir ectly pr opor t ion a l t o
a t t r a ct ion s a n d in ver sely pr opor t ion a l t o cost s (or im peda n ce).
Th u s, a m odel of t h e t r ip d ist r ibu t ion is p r odu ced t h a t a pp r oxim a t es t h e a ctu a l,

em pir ical dist r ibu t ion . Th er e a r e a n u m ber of r ea son s wh y t h is would be u seful - to be able
t o a pply t h e m odel t o a differ en t da t a set fr om wh ich it wa s ca libr a t ed, t o u se t h e m odel for
eva lu a t in g a policy int er ven t ion , or t o us e t h e m odel for for ecas t in g fu t u r e crim e t r ip
dis t r ibu t ion . Bu t , wh a t ever t h e r ea son , it h a s t o be r ea lized t h a t t h e m odel is n ot t h e
obser ved dis t r ibu t ion . Th er e will a lwa ys be a differ en ce bet ween t h e obs er ved dis t r ibu t ion
from which a model is const ru cted a nd t he r esulting predicted distr ibut ion of th e model. It
is u sefu l t o compa r e t h e obs er ved a n d pr edict ed m odel beca u se t h is a llows a t est of th e
validity of t h e imp eda n ce fu n ct ion . But , ra r ely, if ever, will t h e pr edict ed dist r ibut ion be
ide n t ical t o th e em pir ical d ist r ibu t ion.
Anoth er way to th ink of th is is th at th e actu al distr ibut ion of crime tr ips is complex,
r epr esen t in g a la r ge n u m ber of differ en t decis ion s on t h e pa r t of offen der s wh o do n ot
n ecess a r ily use t h e sa m e decision logic. Th e m odel, on t h e oth er h a n d, is a sim ple
a llocat ion on t h e ba sis of t h r ee or , som et im es, four va r ia bles . Alm ost by definit ion , it will
be m u ch sim pler t h a n t h e r ea l dist r ibut ion . Still, t h e sim ple m odel ca n oft en ca pt u r e t h e
m ost imp or t a n t ch a r a ct er ist ics of t h e a ct u a l dist r ibut ion . Hen ce, modelin g ca n be an
14.3
extr em ely u seful an a lyt ica l exer cise t h a t a llows oth er t ypes of ques t ion s t o be as ked t h a t
a r e n ot possible wit h ju st t h e obs er ved dis t r ibu t ion .
Th e G ra v i ty Mo d e l
A m od el t h a t is u su a lly u s ed for t r ip dis t r ibu t ion is t h a t of t h e gravity fun ction , a n

a pp lica t ion of Newt ons fu n da m en t a l law of a t t r a ction (Oppen h eim , 1980; F ield a n d
Ma cGregor, 1987; Or t u za r a n d Willu m sen , 2001). Mu ch of t h e discu ss ion below is a lso
r epea t ed in cha pt er 9 on jou r n ey t o cr im e sin ce th er e is a com m on t h eor et ical ba sis . In t h e
original Newtonian form ulat ion, the a tt ra ction, F, between t wo bodies of respective masses
M 1 an d M 2 , sepa r a t ed by a d ist a n ce D, will be equ a l t o
M1 M2
F = g ----------------- (14.2)
D2
wh er e g is a con st a n t or s calin g fa ctor wh ich en su r es t h a t t h e equ a t ion is ba la n ced in

t er m s of th e m ea su r em en t u n it s (Op pen h eim , 1980). As w e a ll kn ow, of cour se , g is t h e
gr a vit a t ion a l con s t a n t in t h e N ewt on ia n for m u la t ion . Th e n u m er a t or of t h e fu n ct ion is t h e
attraction t er m (or , a lt er n a t ively, t h e a t t r a ct ion of M 2 for M 1 ) wh ile t h e den om in a t or of th e
equ a t ion, d 2 , in dica t es t h a t t h e a t t r a ct ion bet ween t h e t wo bod ies fa lls off a s a fu n ct ion of
t h eir squ ared d is t a n ce. It is a n im p ed a n ce (or r esist a n ce) t er m .
S o c i a l Ap p li c a ti o n s o f t h e G ra v i ty Co n c e p t
Th e gra vit y model ha s been t h e bas is of m a n y app lica t ion s t o h u m a n societies a n d

h a s been a pplied t o social int er a ct ion s sin ce t h e 19 t h cen t u r y. Ra ven st ein (1895) a n d
An d er s son (1897) a p plied t h e con cep t t o t h e a n a lys is of m igr a t ion by a r gu in g t h a t t h e
t en den cy t o m igr a t e bet ween r egion s is in ver sely pr opor t ion a l t o t h e squ a r ed dis t a n ce
bet ween t h e r egions . Reillys la w of r et a il gra vit a t ion (1929) ap plied t h e N ewt onia n
gra vit y model directly an d su ggest ed t h a t r et a il t r a vel bet ween t wo cen t er s would be
pr opor t ion a l t o t h e pr odu ct of t h eir popu la t ion s a n d in ver sely pr opor t ion a l t o t h e squ a r e of
t h e dist a n ce sepa r a t ing t h em :
Pi Pj
I ij = "----------------- (14.3)
D ij2
wh er e I ij is th e int eraction between center s I an d j, P i an d P j ar e the respective populat ions,

D ij is th e dista nce between t hem r aised to th e second power an d " is a ba la n cing cons t a n t .
In t h e m odel, t h e in it ia l popu la t ion, P i , is called a prod u ction wh ile t h e secon d popu la t ion ,
P j, is called an attraction .
St ewa r t (1950) a n d Zipf (1949) a pplied t h e con cept t o a var iety of ph en om en a

(m igr a t ion , fr eigh t t r a ffic, in for m a t ion ) u sin g a sim plified for m of t h e gr a vit y equ a t ion
14.4
Pi Pj
I ij = K ----------------- (14.4)
D ij
wh er e t h e t er m s a r e a s in equa t ion 14.3 but t h e expon en t of dist a n ce is on ly 1. Given a

pa r t icula r pa t t er n of in t er a ction for a n y t ype of goods , ser vice or h u m a n a ctivit y, a n
opt im a l loca t ion of fa cilities sh ould be s olvable.
In t h e St ewa r t /Zipf fr a m ewor k , t h e t wo Ps wer e both popu la t ion sizes . H owever , in

m oder n u se, it is n ot n ecess a r y for t h e pr odu ctions a n d a t t r a ctions t o be iden t ical u n it s
(e.g., P i cou ld be popu lat ion wh ile P j cou ld be em ploym en t ).
Tri ps as In te ra ct io n s
It sh ou ld be obviou s t h a t t h is in t er a ct ion equ a t ion ca n be a pplied t o t r ip s fr om on e

a r ea (zon e) t o a n ot h er . Ch a n gin g t h e sym bols sligh t ly, t h e t ot a l volu m e of t r ip s fr om a
pa r t icu la r or igin zon e, i, t o a sin gle loca t ion , j, is dir ect ly pr opor t ion a l t o t h e pr odu ct of th e
p rod uct ion s a t i a n d t h e a t t r a ct ion s a t j, a n d in ver sely pr opor t ion t o t h e im peda n ce (or
cost ) of t r a vel bet ween t h e t wo zon es
" P i $ Aj
T ij = --------------- (14.5)
D ij
wh er e P i a r e t h e pr odu ction s for zone I, Aj a r e t h e a t t r a ction s zon e j, " is a pr odu ction

con st a n t , $ is an at tr action const an t, and D ij is t h e im peda n ce (cost ) of t r a vel bet ween zon e
ii a n d zon e j.
Over time, the concept ha s been genera lized an d applied to ma ny different types of
t r a vel beha vior . For exam ple, Hu ff (1963) a pplied t h e con cept t o r et a il t r a de bet ween
zon es in a n u r ba n a r ea u sin g t h e gen er a l for m of
S j8
Aij = " ----------------- (14.6)
D ijD
wh er e Aij is t h e n u m ber of pu r ch a ses in loca t ion j by r esid en t s of loca t ion I, S j is t h e

a t t r a ctiven es s of zon e j (e.g., squ a r e foota ge of r et a il sp a ce), D ij is t h e dist a n ce bet ween
zones I a n d j, " is a con st a n t , 8 is t h e expon en t of S j, a n d D is t h e exp on en t of dis t a n ce
(Boss a r d, 1993). Dij-D is som et im es ca lled a n in v erse d ist an ce fu n ct ion . Th is differ s fr om
t h e t r a dit ion a l gr a vit y fu n ct ion by a llowin g t h e exp on en t s of t h e pr odu ct ion fr om loca t ion I,
t h e a t t r a ct ion fr om loca t ion j, a n d t h e dist a n ce bet ween zones t o var y.
E qu a t ion 14.6 is a sin gle con st ra in t m odel in t h a t only t h e a t t r a ct iven ess of a

com m er cia l zon e is con st r a in ed , t h a t is th e s um of a ll a t t r a ct ion s for j m u s t equ a l t h e t ot a l
14.5
a t t r a ction in t h e r egion. Aga in , it can be gen er a lized t o all zon es by, first , est im a t in g t h e
t ota l t r ips gen er a t ed from one zone, i, t o an oth er zone, j,
P i 8 AjJ
T ij = " ----------------- (14.7)
D ijD
wh er e T ij is t h e in t er a ction bet ween t wo locat ions (or zones), P i is pr od uct ion s of t r ip s fr om

zone I, Aj is t h e a t t r a ctiven es s of zon e j, D ij is t h e dist a n ce bet ween zones I a n d j, 8 is t h e
exp on en t of P i , J is t h e expon en t of Aj, D is th e exponent of dista nce, an d " is a con st a n t .
Secon d, t h e t ota l n u m ber of t r ips gen er a t ed by a locat ion , I, t o all dest in a t ion s is
obt a in ed by s u m m in g over a ll d es t in a t ion locat ion s, j:
T i = " P i 8 G (AjJ /D ijD ) (14.8)
a n d gener a lizing t h is t o a ll zon es, we get:
" P i 8 $ AjJ
T ij = ------------ (14.9)
D ijD
wh er e " is a con st a n t for t h e pr odu ction s, P i 8 , but $ is a con st a n t for t h e a t t r a ction s, AjJ .
Th is t ype of fu n ct ion is ca lled a d ou ble con st ra in t m odel beca u se t h e equa t ion h a s t o be
con s t r a in ed by t h e n u m ber of u n it s in bot h t h e or igin a n d d es t in a t ion loca t ion s ; t h a t is , t h e
s um of P i over a ll loca t ion s m u st be equ a l t o t h e t ot a l n u m ber of p r od u ct ion s wh ile t h e s u m
of Aj over a ll loca t ion s m u st be equ a l t o t h e t ot a l n u m ber of a t t r a ct ion s. Ad ju st m en t s a r e
u su a lly requ ir ed t o ha ve t h e su m of in dividu a l pr odu ctions a n d a t t r a ctions equ a l t h e t ota ls
(u su a lly est im a t ed in depen den t ly).
N e g a t i v e E x p o n e n t i a l D i s t an c e F u n c t i o n
On e of t h e pr oblems with t h e t r a dit ion a l gr a vit y for m u lat ion is in t h e m ea su r em en t

of t r a vel im peda n ce (or cost ). For loca t ions sepa r a t ed by sizea ble d ist a n ces in sp a ce, th e
gr a vit y for m u la t ion ca n wor k pr oper ly. H owever , a s t h e dis t a n ce bet ween loca t ion s
decr ea ses, t h e den om in a t or a ppr oa ch es in fin it y. Con sequ en t ly, a n a lt er n a t ive exp r ession
for t h e in t er a ct ion u ses t h e n ega t ive exp on en t ia l fu n ct ion (H ger st r a n d, 1957; Wils on ,
1970).
T ji = Aj$ e (-"D ij ) (14.10)
wh er e T ji is t h e a t t r a ction of loca t ion j for r es ide n t s of locat ion I , Aj is th e att ra ctiveness of

loca t ion j, D ij is th e dista nce between locat ions i and j, $ is t h e expon en t of S j, a n d e is t h e
ba se of th e n a t u r a l loga r it h m (i.e ., 2.7183...). Der ived fr om pr in cip les of en tropy
14.6
m axim ization, th e lat ter pa rt of th e equat ion is a negative exponent ial fun ction t ha t h as a
m a xim u m va lu e of 1 (i.e., e -0 = 1) (Wilson , 1970). Th is h a s t h e a dva n t a ge of m a k in g t h e
equ a t ion m or e st a ble for in t er a ct ion s bet ween loca t ion s t h a t a r e clos e t oget h er . F or
exa m ple, Cliff an d H a gget t (1988) us ed a n ega t ive expon en t ia l gra vit y-t ype m odel t o
descr ibe t h e diffu sion of m ea sle s in t o th e U n it ed St a t es from Ca n a da a n d Mexico. It h a s
a lso been a r gu ed t h a t t h e n ega t ive expon en t ia l fu n ction gen er a lly gives a bet t er fit t o
u r ba n t r a vel pa t t er n s, p a r t icula r ly by a u t omobile (Foot, 1981; Boss a r d, 1993;). F igu r e 14.2
sh ows a t yp ica l n ega t ive exp on en t ia l fu n ct ion a n d on e r ecom m en ded for h om e-ba sed wor k
t r ip s by t h e Tr a n spor t a t ion Resea r ch Boa r d a s a defa u lt va lu e (NCH RP , 1995).
Note t h a t by m ovin g t h e dist a n ce te r m t o th e n u m er a t or, st r ictly spea k in g it n o

lon ger is a n im peda n ce te r m sin ce imp eda n ce incr ea ses wit h dis t a n ce. Rat h er it is a
d iscou n t fa ct or (or disincentive); th e int eraction is discoun ted with dista nce. Neverth eless,
t h e t er m im peda n ce is st ill u sed, p r im a r ily for h is t or ica l r ea son s.
Th er e a r e ot h er dis t a n ce fu n ct ion s, a s well. Ch a pt er 9 exp lor ed som e of t h ese. F or

exam ple, we ar e find ing t h a t , for cr ime t r ips, th ese oth er fu n ct ion s m a y produ ce bet t er
r esu lts t h a n t h e n egat ive expon en t ial (e.g., t h e lognorm a l), prim a r ily beca u se m a n y cr imes
ar e comm itted at short -to-moderat e dista nces.
Tra v e l Im p e d a n c e
On e of t h e biggest a dva n ces in t h is t ype of model h a s been t o in crea se t h e flexibility

of th e denomina tor. In th e tra ditiona l gra vity model, th e denomina tor is dista nce. This is
a pr oxy for a discoun t factor (or cost ); t h e fa r t h er t wo zon es a r e fr om ea ch ot h er , th e less
lik ely t h er e is t o be in t er a ction bet ween t h em , a ll oth er t h in gs bein g equ a l. Conver se ly, t h e
close r t wo zon es a r e, t h e m ore likely t h er e is t o be in t er a ction , a ll oth er t h in gs bein g equ a l.
Trave l Time
It h a s been r ea lized, however, th a t dist a n ce is on ly a n a ppr oxim a t ion t o imp eda n ce.
In r ea l tr a vel, tr a vel t ime is a m u ch bet t er ind ica t or of t h e cost of t r a vel in t h a t tim e va r ies
by t h e t im e of da y, da y of week , an d ot h er fact ors . For exa m ple, t r a vel a cross t own in a n y
m et r opolita n a r ea is gener a lly a lot ea sier a t 3 in t h e m or n ing, sa y, t h a n a t t h e pea k
a ft er n oon r u sh per iod. The d iffer en ce in t r a vel t ime can var y as m u ch a s t wo-t o-t h r ee
t im es bet ween pea k a n d off-pea k h ou r s. U sin g on ly dis t a n ce, h owever , t h ese va r ia t ion s a r e
n ever picke d u p be cau se t h e dist a n ce bet ween locat ion s is in va r ia n t .
Th is r ea liza t ion h a s led t o t h e concep t of t ra vel im p ed a n ce which, in t ur n, has led to

t h e con cep t of trav el cost. Im peda n ce is t h e r esist a n ce (or discou n t ing) in t r a vel bet ween
t wo zon es. Usin g tr a vel t ime a s a n imp eda n ce var iable, th e lon ger it t a kes t o t r a vel
bet ween t wo zon es , t h e les s lik ely t h er e will be in t er a ction bet ween t h em , a ll oth er t h in gs
being equ a l. Con vers ely, a sh or t er t r a vel t ime lea ds t o grea t er int er a ct ion bet ween zon es,
a ga in , all ot h er t h in gs bein g equ a l. Sim ila r ly, a t r a vel r out e t h a t sh ort en s t r a vel t im e will
14.7
Figure 14.2:
Default Home-Based Work Trip Impedance

(Source: National Cooperative Highway Research Program 365, 1995)
0.0020
Relative impedance
0.0015
0.0010
0.0005
0.0000
0 4 8 12 16 20 24 28
Distance (miles)
gen er a lly be s elect ed over one t h a t t a k es longer even if th e first one is lon ger in dis t a n ce.
F or exam ple, it s been docu m en t ed t h a t people will ch a n ge wor k loca t ion s t h a t a r e fa r t h er
from t h eir h ome if t r a velin g t o th e n ew work locat ion t a k es less t im e (e.g., t r a velin g in t h e
opposit e dir ect ion t o th e bu lk of t r a ffic; Wa chs , Ta ylor, Levin e & On g, 1993).
If t r a vel t im e is a crit ical com pon en t of t r a vel, wh y t h en don t offend er s com m it

m or e cr imes a t , say, 3 in t h e m or n ing t h a n a t t h e pea k a ft er n oon t r a vel t imes ? Since th e
imp eda n ce is less a t 3 in t h e m or n ing t h a n a t , say, 5 in t h e a ft er n oon , wou ldn t t h e m odel
pr edict m or e t r ip s occu r r in g in t h e ea r ly m or n in g h ou r s t h a n a ct u a lly occu r in t h ose h ou r s?
Th e a n swe r h a s t o do wit h t h e n u m er a t or of th e gr a vit y equ a t ion a n d n ot ju st t h e
den om ina t or . At 3 in t h e m or n ing, yes, it is ea sier t o t r a vel bet ween t wo loca t ion s, at leas t
by per sona l au t om obile (n ot by bus or t r a in w h en t h ose s er vices a r e less fr equ en t ). But t h e
a t t r a ct ion side of t h e equa t ion is also less st r on g at 3 in t h e m or n ing. For a st r eet r obber ,
t h er e a r e fewer pot en t ia l vict im s on t h e st r eet a t 3 in t h e m orn in g t h a n in t h e la t e
a ft er n oon . For a r esiden t ial bu r gla r , th er e is m or e likely to be som eon e a t h om e wh ile t h ey
bu r gle a t n igh t t h a n in t h e a ft er n oon . Th e t r a vel t im e com pon en t is on ly on e dim en sion of
t h e likelih ood of t r a vel bet ween t wo loca t ion s. The d ist r ibut ion of opport u n ities a n d oth er
cost s ca n a lt er t h e lik elih ood cons ide r a bly.
Never t h eless, sh ift ing t o a n imp eda n ce fu n ct ion a llows a t r a vel m odel t o bet t er
r eplica t e a ct u a l t r a vel con dit ion s. Most t r a vel d em a n d m odels u sed by t r a n spor t a t ion
pla n n er s u se a n im peda n ce fu n ct ion , r a t h er t h a n a dis t a n ce fu n ct ion . 1 Dist a n ce wou ld only
be m ea n ingfu l if t h e st a n da r ds wer e inva r ian t with r espect t o t ime (e.g., a m odel ca lcu lat ed
over a n en t ir e yea r , 24 h ou r s a da y). As will be dem on st r a t ed in ch a pt er 16 on n et wor k
a ssign m en t , a t r a vel t im e ca lcu la t ion lea ds t o a ver y d iffer en t n et wor k a lloca t ion t h a n a
dist a n ce ca lcu lat ion . For exam ple, if dist a n ce is us ed a s a n imp eda n ce var iable, th en t h e
sh ort est t r ips will r a r ely t a k e t h e fre ewa ys becau se t r a vel t o an d fr om a free wa y usu a lly
ma kes a t rip longer th an a direct r out e between a n origin and a destinat ion. But as m ost
people u n der st a n d, ta kin g a freewa y to tr a vel a sizeable dist a n ce is us u a lly a lot quicker
t h a n t r a ver sin g a n u r ba n a r t er ia l sys t em wit h m a n y t r a ffic light s, s t op sign s, crossin g
pedest r ia n s, cr oss t r a ffic fr om p a r k in g lots a n d s h oppin g m a lls, a n d ot h er u r ba n obst a cles.
Toda y, t h e u se of dis t a n ce in t r a vel d em a n d m odelin g h a s vir t u a lly been dr opped by m ost
tr an sport at ion plann ers.
Tr a v e l Co s t
An even bet t er con cep t of im p ed an ce is th a t of trav el cost (somet imes ca lled

gen eralized cost) wh ich incor pora t es r ea l an d per ceived cost s of t r a vel bet ween t wo
loca t ion s . Tr a vel t im e is on e com p on en t of t r a vel cos t in t h a t t h er e is a n im p licit cos t t o t h e
t r ip (e.g., an h our ly wa ge or p r ice as sign ed). In t h is ca se, t wo differ en t in dividu a ls will
va lu e t h e t im e for a t r ip differ en t ly depen din g on t h eir h ou r ly wa ge. F or exa m ple, for a n
in divid u a l wh o pr ices h is /h er t r a vel a t $100 a n h ou r , t h e per m in u t e cost is $1.67. F or
a n oth er in divid u a l wh o pr ices h is/h er t r a vel a t $12 a n h our , t h e per m in u t e cost is 20.
Th ese r elat ive pr ices a ssign ed t o t r a vel will su bst a n t ially affect ind ividu a l ch oices in t r a vel
m odes a n d r ou t es. For in st a n ce, th ese t wo h ypot h et ica l in dividu a ls will pr oba bly u se a
14.9
differen t t r a vel m ode in gett ing fr om an air por t t o a h ot el on a t r ip; t h e for m er will
pr obably t a k e a t a xi wh er ea s t h e la t t er will pr obably t a k e a bu s or t r a in (if a va ila ble).
Bu t cos t in volves ot h er d im en s ion s t h a t n eed t o be con s id er ed . Th er e a r e r ea l

oper a t ing cost s in t h e u se of a vehicle - fu el, oil, m a int en a n ce, in su r a n ce. Man y tr a vel
st u dies h a ve su ggest ed t h a t dr iver s in cor por a t e t h ese cost s a s p a r t of t h eir im plicit h our ly
t r a vel p r ice (Or t u za r a n d Willu m sen , 2001; 323-327). Bu t , t h er e a r e a ls o r ea l, ou t -of-
pocket cost s su ch a s pa r k in g or t oll cost s. P a r k in g is pa r t icu la r ly a m a jor exp en se for
int r a -u r ba n dr iving beh a vior . In m a n y built-up bus ines s dist r ict s, pa r kin g cost s can be
considerable, for example as m uch a s $40 a day in m ajor m etropolita n center s. In most
bu sy comm er cial a r ea s, t h er e a r e some pa r k in g cost s, if only a t on-st r eet pa r k in g m et er s.
Th u s, a t r a vel cos t m odel n eeds t o in cor por a t e t h ese r ea l cos t s a s t h e ou t -of-pocket cost s
m a y overwh elm t h e imp licit va lue of t h e t r a vel t ime. F or exam ple, an offen der wh o lives
10 m in u t es fr om t h e d own t own a r ea by ca r wou ld pr oba bly n ot dr ive in t o t h e d own t own t o
com m it a r obber y since th a t ind ividu a l will ha ve to bear t h e pr ice of pa r kin g. Ther e a r e
lot s of well k n own s t or ies t h a t cir cu la t e a bou t ba n k r obber s wh o a r e ca u gh t beca u s e t h ey
in cu r p a r kin g t ick et s wh ile com m it t in g t h eir cr im e. H ow oft en t h is h a s occu r r ed is n ot
k n own fr om a n y s t u dy t h a t Im a wa r e, bu t t h e st or y lin e is cogn iza n t of th e a ct u a l cos t s of
t r a vel t h a t m u st be in cur r ed a s p a r t of t r a vel.
In a ddit ion t o r ea l cost s a r e per ceived cost s. For t r a n sit u ser s, pa r t icu lar ly, th ese
per ceived cost s a ffect t h e ea se a n d t ime of t r a vel. One of t h e st a n da r d qu est ion s in t r a vel
su r veys for t r a n sit u ser s is t h e t ime it t a kes t o wa lk from t h eir h om e t o t h e n ea r est bus
st op or in t r a -ur ba n r a il syst em (if ava ila ble) an d fr om t h e la st t r a n sit st op t o th eir fin a l
destinat ion; th e longer it t ak es to access the tr an sit system, th e less likely an individua l
will u se it . Similar ly, tr a n sfers bet ween bus es or t r a ins decrea se t h e likelih ood of t r a vel by
t h a t m ode, a lm ost in pr opor t ion t o th e n u m ber of t r a n sfer s. Th e r ea son is t h e difficult y in
get t in g ou t of on e bu s or t r a in a n d in t o a n ot h er . Bu t , t h e t im e bet ween t r a in s a dd s a n
imp licit t r a vel cost ; t h e lon ger t h e wa it bet ween bus es, th e less likely th a t m ode will be
u sed by t r a veler s. In sh ort , ea se of a ccess a n d con ven ien ce ar e posit ive in cent ives in u sin g
a m ode or a r out e wh ile difficult y in a ccessin g it, lack of con ven ien ce, an d even fear of bein g
vuln er a ble to cr ime will decr ea se t h e likelih ood of u sin g th a t m ode or r ou t e. 2
If t h e con cept is exp a n ded t o t h a t of a n offen der , t h er e a r e ot h er per ceived cost s t h a t

m igh t a ffect t r a vel. On e obviou s on e is t h e lik elih ood of be in g ca u gh t . It m a y be ea sy for
on e offen der t o t r a vel t o a u pscale, high visibility sh oppin g ar ea , but if t h er e a r e m a n y
police a n d s ecur it y gu a r ds a r oun d, t h e in divid u a l is m ore likely t o be cau gh t . H en ce, th a t
lik elih ood (or , m or e a ccu r a t ely, a n a ss u m pt ion t h a t t h e offen d er m a k es a bou t t h a t
lik elih ood sin ce h e/sh e doesn t r ea lly k n ow wh a t is t h e r ea l lik elih ood) is lia ble t o a ffect t h e
ch oice of a dest ina t ion a n d, possibly, even a r ou t e.
An ot h er per cep t u a l com pon en t a ffect in g a lik ely ch oice is t h e r elia bilit y of t h e
t r a n sport a t ion m ode. Man y offen der s a r e poor a n d dont h a ve expen sive, well m a int a ined
veh icles. If th e veh icle is n ot ca pa ble of h igher sp eed s or is even like ly to brea k down wh ile
a n offen ce is bein g com m it t ed, t h a t veh icle is n ot lia ble t o be u sed in m a k in g a t r ip or t h e
ch oice of dest ina t ion m a y be alt er ed. It is well kn own t h a t m a n y offen der s st ea l veh icles
14.10
for u se in a cr im e. F ea r s a bou t n ot bein g iden t ified a r e clea r ly a m a jor fa ct or in t h ose
decis ion s, bu t t h e r elia bilit y of t h eir own veh icles m a y a ls o be a fa ct or .
Th u s, in sh ort , a m ore r ea list ic m odel of t h e in cent ive or d isin cent ive t o ma k e a t r ip

bet ween t wo loca t ion s r equ ir es a com plex fu n ct ion t h a t weigh t s a n u m ber of fa ct or s
a ffect in g t h e cost of t r a vel - t h e t r a vel t im e, im plicit oper a t in g cos t s, ou t -of-pocket cost s,
a n d p er ceived cost s. Ma n y t r a vel dem a n d m odels u sed by Met r opolit a n P la n n in g
Or ga n iza t ion s u se su ch a fu n ct ion , u su a lly u n der t h e la bel of gen er a lized cost . Th e m or e
com p lex t h e p ricin g s tr u ct u r e for pa r k in g a n d t r a vel wit h in a m et r op olit a n a r ea , t h e m or e
lik ely a gen er a lized cost fu n ct ion will p r ovide a r ea list ic m odel of t r ip dis t r ibu t ion .
Trave l Utility
Th e fin a l con cep t t h a t is in t r odu ced in defin in g im peda n ce is t h a t of travel utility.

Ut ilit y is a n in divid u a l con cept , r a t h er t h a n a zon a l on e. Also, it is t h e flip sid e of cost
(i.e., h igh er cost is as sociat ed wit h less u t ilit y). A gener a lized cost fu n ct ion ca lcu lat es t h e
object ive a n d a vera ge perceived cost s of t r a vel bet ween t wo zon es. But t h e u t ilit y of t r a vel
for a n in d ivid u a l is a fu n ct ion of bot h t h os e r ea l cos t s a n d a n u m ber of in d ivid u a l
ch a r a ct er is t ics t h a t affect t h e va lu e p la ced on t h a t t r a vel. Th u s , t wo in d ivid u a ls livin g in
t h e sa m e zon e (per h a ps even livin g next door t o ea ch ot h er ) wh o t r a vel t o t h e sa m e
dest ina t ion loca t ion m a y pr ice t h eir t r ip very differ en t ly. Aside from incom e differ en ces
wh ich a ffect s t h e a vera ge hour ly wa ge, th er e m a y be differ en ces du e t o con venien ce,
a t t r a ct iven ess, or a h ost of ot h er fa ct or s. Ot h er fa ct or s a r e m or e id ios yn cr a t ic. F or
exam ple, a t r ip by a gan g mem ber int o a n ot h er gan gs t u r f m igh t be expect ed t o increa se
t h e per ceived cost s t o t h e in divid u a l of t r a velin g t o t h a t loca t ion , a bove a n d beyon d a n y
objective cost fact ors . Alt er n a t ively, a t r ip t o a loca t ion wh er e a close frien d or r ela t ive is
loca t ed m igh t decrea se t h e per ceived cost of t r a vel t o t h a t zon e. In oth er wor ds, t h er e a r e
both object ive cost s a s w ell a s s u bjective cost s in t r a vel bet ween t wo zon es .
Th e con cept of u t ilit y ma y be less u seful for cr ime a n a lysis t h a n for gener a l tr a vel
beh a vior . For on e t h ing, since th e con cept is individua l, it can on ly be ident ified by
in dividu a l su r veys (Domen cich a n d McFa dd en , 1975). For cr im e a n a lysis , t h is m a k es it
vir t u a lly imposs ible t o us e sin ce it is ver y difficult t o in t er view offend er s, a t lea st in t h e
Un it ed St a t es. In a ddit ion , t h e m a t h em a t ics r equ ir ed for a r t icu la t in g a u t ilit y fu n ct ion a r e
difficu lt sin ce u t ilit y fu n ct ion s h a ve a ver y com plex for m , u su a lly in volvin g t h e bin om ia l or
m u lt in omia l logit fun ction wit h a Weibu ll er r or t er m . In t h e n ext cha pt er , br ief me n t ion is
m a de of t h is t ype of m odel.
But, for completeness sake, we need to un dersta nd t ha t t he likelihood or

dis in cen t ive t o t r a vel bet ween t wo loca t ion s is a fu n ct ion of in divid u a l ch a r a ct er is t ics a s
well a s object ive t r a vel cost com pon en t s.
Im p e d a n c e F u n c t i on
Th u s, for a zon a l type m odel, we ca n leave t h e gra vit y fu n ct ion a s a gener a lized
imp eda n ce fu n ct ion . For t r a vel bet ween a n y on e zon e a n d a ll ot h er zon es, we ha ve:
14.11
T i = " P i 8 G (AjJ/I ij) (14.11)
wh er e t h e n u m ber of t r ip s fr om zon e i t o a ll ot h er zon es, j, is a fu n ct ion of t h e pr odu ct ion s

a t zon e i a n d t h e r ela t ive a t t r a ction of a n y on e zone, j, t o th e im peda n ce of th a t zon e for i,
I ij. Th e im peda n ce fu n ction , I ij , is some d eclining fun ct ion of cost for t r a vel bet ween t wo
zon es. It does not h a ve to be an y par t icu lar for m a n d can be (a n d u su a lly is) a n on -linea r
fun ction. Th e cost s ca n be in t er m s of dis t a n ce, tr a vel t im e, sp eed (which is con ver t ed in t o
t r a vel t im e) or gen er a l cost s. Th e gr ea t er t h e sepa r a t ion bet ween t wo zon es (i.e., th e
h igher t h e im peda n ce), t h e les s lik ely t h er e will be a t r ip bet ween t h em . Gen er a lizing t h is
t o a ll zon es, we get:
T ij = " P i 8 $ AjJ I ij (14.12)
wh er e P i is t h e pr odu ction cap a city of zon e I, Aj is t h e a t t r a ction of zone j, I ij, is a

gen er a lized fu n ct ion t h a t dis cou n t s t h e in t er a ct ion wit h in cr ea s in g s ep a r a t ion in d is t a n ce,
t im e, or cost , " an d $ a r e con st a n t s t h a t a r e a pplied t o t h e pr odu ct ion s a n d a t t r a ct ion s
respectively, an d 8 an d J a r e fin e t u n in g exp on en t s of t h e pr odu ct ion s a n d a t t r a ct ion s
r espectively. This is t h e gra vit y fu n ct ion t h a t we will est ima t e in t h e Crim eS tat t r ip
dis t r ibu t ion m odel.
Alterna tive Mode ls: Inte rven ing Oppo rtun ities
Th er e a r e a lt er n a t ive a lloca t ion s pr ocedu r es t o t h e gr a vit y m odel. On e well kn own

on e is t h a t of int erven ing opportun ities. St ouffer (1940) modified t h e sim ple gr a vit y
fu n ct ion by a r gu in g t h a t t h e a t t r a ct ion bet ween t wo loca t ion s wa s a fu n ct ion n ot on ly of t h e
ch a r a ct er ist ics of t h e r elat ive a t t r a ct ion s of t wo loca t ion s, but of int er venin g opport u n ities
bet ween t h e loca t ion s. H is h ypot h esis ..ass u m es t h a t t h er e is n o necessa r y r ela t ion sh ip
bet ween m obility a n d d ist a n ce... t h a t t h e n u m ber of per son s going a given dis t a n ce is
directly proport iona l to th e num ber of opport un ities at t ha t dista nce and inversely
pr opor t ion a l t o th e n u m ber of in t er ven in g opp ort u n it ies (St ouffer , 1940, p . 846). This
m odel wa s u se d in t h e 1940s t o expla in in t er st a t e a n d in t er -coun t y m igr a t ion (Br igh t a n d
Th om a s, 1941; Isbell, 1944; Isa r d, 1979). Usin g th e gra vit y type for m u lat ion , th is ca n be
written as:
S j$
Aji = " ----------------- (14.13)
G(S k >) D ij 8
wh er e Aji is t h e a t t r a ction of loca t ion j by r es ide n t s of locat ion I , S j is th e att ra ctiveness of

zone j, S k is t h e a t t r a ct iven ess of a ll ot h er loca t ion s t h a t a r e interm ediate in d is t a n ce
bet ween loca t ions I a n d j, D ij is th e dista nce between zones I an d j, $ is t h e expon en t of S j, >
is t h e expon en t of S k , a n d 8 is t h e exponen t of dis t a n ce. Wh ile t h e in t er ven in g
opport u n it ies a r e im plicit in equ a t ion 14.7 in t h e exp onen t s, $ an d 8 , a n d coefficient , ",
equ a t ion 14.13 ma kes t h e int er venin g opport u n ities explicit. Th e imp or t a n ce of t h e con cept
14.12
is t h a t t r a vel bet ween t wo loca t ion s becom es a com plex fu n ct ion of t h e spa t ial en vir on m en t
of n ea r by a r ea s a n d n ot ju st of t h e t wo loca t ion s.
In pr a ctice, in sp it e of it s m ore in t u it ive t h eore t ical m odel, t h e in t er ven in g

opport u n ities m odel does n ot imp r ove pred ict ion m u ch beyon d t h a t of t h e gra vit y model
sin ce it in clu des t h e a t t r a ct ion s a ssocia t ed wit h t h e dest in a t ion zon es. Also, it is a m or e
difficu lt m odel t o est im a t e sin ce t h e a t t r a ct ion s of a ll ot h er zon es m u st be con sid er ed for
ea ch zon e pa ir (or igin -dest in a t ion com bin a t ion ). Con sequ en t ly, it is r a r ely u sed in a ct u a l
pr a ctice (Or t u za r a n d Willu m se n , 2001; Zh a o et a l, 2001).
An ot h er a lt er n a t ive m et h od wa s con d u ct ed by P or oja n (2000) in a p plyin g t h e

gr a vit y m od el t o in t er n a t ion a l t r a de flow. H e a dd ed a s pa t ia l a u t ocor r ela t ion com p on en t in
a ddit ion t o im peda n ce a n d obt a in ed a sligh t ly bet t er fit t h a n t h e pu r e gr a vit y fu n ct ion .
H owever, wh et h er t h is a ppr oa ch wou ld imp r ove th e fitt ing of int r a -r egion a l cr ime t r a vel
pa t t er n s in st ill u n kn own . Nevert h eless, th is a n d oth er a ppr oa ch es m igh t imp r ove th e
pr edicta bilit y of a gr a vit y fu n ction for in t r a -ur ba n crim e t r a vel.
Me t h o d o f E s ti m a t i o n
Th e Crim eS tat t r ip d ist r ibu t ion m odel im plem en t s equ a t ion 14.12. Th e specific
det a ils a r e discu ssed below, but t h e model is iter a t ive. The st eps a r e as follows:
1.Depend ing on wh et h er a sin gly con st r a ined or doubly con st r a ined m odel is to be
est im a t ed, it st a r t s wit h a in it ia l gu ess of t h e va lu es for " or $ (or both for a dou bly
const ra ined model). Table 14.1 illust ra tes th e thr ee models.
Ta ble 14 .1
Th r e e Me t h o d s o f Co n s t r a i n i n g t h e G ra v i t y Mo d e l
S in g le c on s tra in t
Cons t r a in origin s
Ti j = " P i 8 A jJ I i j
Con st r a in dest in a t ion s
Ti j = P i 8 $ AjJ I i j
D o ub le c on s tra in t
Con st r a in bot h or igin s a n d dest in a t ion s
Ti j = " P i 8 $ AjJ I i j
14.13
2. Th e r out in e pr oceed s t o est im a t e t h e va lu e for ea ch cell in t h e origin -
dest in a t ion m a t r ix (see figu r e 14.1 a bove) u sin g t h e exist in g est im a t es for "
an d $.
3. Th e r ou t ine t h en su m s t h e r ows a n d colum n s in t h e m a t r ix. Then ,

depen din g on wh et h er a sin gle- or double-con st r a int m odel is to be est ima t ed
a n d, if a s in gle-cons t r a in t , wh et h er origin s or dest in a t ion s a r e t o be held
con s t a n t , it t h en ca lcu la t es t h e r a t io of t h e s u m m ed va lu e (r ow or colu m n or
bot h ) t o t h e in it ia l r ow or colu m n s u m . Th e in ver s e of t h a t r a t io is t h e
su bsequ en t est im a t e for " or $ (or bot h for a dou ble-const r a in ed m odel).
4. Th e r ou t in e r epea t s st eps 2 a n d 3 u n t il t h e ch a n ges fr om on e it er a t ion t o t h e

n ext a r e ver y sm a ll.
5. Th e la st es t im a t e of " or $ (or bot h for a double-con st r a ined m odel) is t a ken

as t he final values of th ese par am eters.
6. On ce t h e pa r a m et er s h a ve been est im a t ed, t h e m odel ca n be a pplied t o t h e

ca libr a t ion da t a set or t o a n ot h er da t a set . N ot e t h a t t he p ar a m et er s a r e r ow
or colu m n sp ecific (or both ). Th a t is, in t h e con st r a in origin s m odel, t h er e is
a sepa r a t e coefficien t for ea ch r ow. In t h e con st r a in dest in a t ion s m odel,
t h er e is a sepa r a t e coefficien t for ea ch colu m n . In t h e con st r a in both origin s
a n d d est in a t ion s, t h er e is a sepa r a t e coefficien t for ea ch cell (r ow-colu m n
com bin a t ion ).
7. A com p a r is on ca n be m a d e bet ween t h e obs er ved dis t r ibu t ion a n d t h e

pr edict ed (modeled) dist r ibut ion . Beca u se m ost or igin-dest ina t ion m a t r ices
a r e ver y lar ge, t h e va st m a jor it y of cells w ill ha ve zer o in t h em . Th u s, a chi-
squ a r e t est wou ld be ina ppr opr iat e. Ins t ea d, a com pa r ison of t h e trip length
dis t r ibu t ion is m a de u sin g t wo differ en t st a t is t ics - a coin cid en ce r a t io a n d
t h e Kom ologor ov-Sm irn ov Two-sa m ple sta t istic. Det a ils a r e pr ovided below.
Cr i m eS t a t III Tri p D is tri bu t io n Ro u ti n e s
Next , we exa m ine t h e a ct u a l tools t h a t a r e a vailable in t h e Crim eS tat t r ip

dis t r ibu t ion m odu le. Th e t ools a r e illu st r a t ed wit h exa m ple s fr om Ba lt im ore Coun t y.
Th e Crim eS tat t r ip dist r ibut ion m odu le in clud es one set u p screen a n d five rout ines
t h a t im plem en t t h e m odel:
1. Ca lc u l a te o b s e r v e d o r ig i n -d e s t i n a t io n d i s t ri b u t io n . If th er e is a file
a va ila ble w it h t h e coord in a t es for in divid u a l origin s a n d d es t in a t ions (e.g.,
a n a r r est r ecor d), t h is r ou t in e will ca lcu la t e t h e em pir ica l t r ip dis t r ibu t ion
m a t r ix;
14.14
2. Ca li b ra t e i m p e d a n c e f u n c t i o n . If th er e is a file a va ila ble wit h t h e
coor din a t es for in divid ua l or igin s a n d d es t in a t ion s, t h is rou t in e will ca libr a t e
a n em pir ica l im peda n ce fu n ct ion .
3. S e tu p ori gi n-d e st in a ti on m o de l. This screen a llows t h e u ser t o defin e t h e

pa r a m et er s of a t r ip dist r ibut ion (or igin-dest ina t ion ) m odel with eith er a
m a t h em a t ica l or em pir ica l im peda n ce fu n ct ion .
4. Ca li bra te o ri gi n-d e st in a ti on m o de l. This rout ine calibrat es the

pa r a m et er s of th e t r ip d ist r ibu t ion m odel (equ a t ion 14.12) us in g t h e
pa r a m et er s defined on t h e set u p pa ge.
5. Ap ply pre d ic te d ori gi n-d e st in a ti on m o de l. This rout ine applies th e

est ima t ed pa r a m et er s t o a da t a set . The da t a set ca n be eith er t h e
ca libra t ion file or a n ot h er file.
6. Co m p a r e o b s e r v e d a n d p re d i c t e d o ri g i n -d e s t i n a t io n t r ip le n g t h s .
Th is r out in e com pa r es t h e t r ip len gt h s fr om t h e obser ved (em pir ical) tr ip
distr ibut ion with t ha t pr edicted by th e model. Compa rison a re ma de
gr a ph ically, by a coin ciden ce ra t io, by t h e Kom ologorov-Sm ir n ov Two-Sa m ple
test, an d by a Chi squa re test on t he most frequent tr ip links.
E a ch of t h ese r ou t ines a r e described in det a il below. Figur e 14.3 shows a screen

sh ot of t h e t r ip dist r ibut ion m odu le.
D e s c r ib e Or ig i n -D e s t i n a ti o n Tr ip s
An em p ir ica l d es cr ip tion of t h e a ct u a l t r ip dis t ribu t ion m a t r ix ca n be m a de if t h er e

is a da t a set t h a t inclu des in dividu a l or igin a n d des t ina t ion loca t ion s. The u ser defin es t h e
or igin loca t ion a n d t h e d es t in a t ion loca t ion for ea ch r ecor d a n d a set of zon es fr om wh ich t o
com pa r e t h e in dividu a l or igins a n d d est in a t ion s. Th e r out in e m a t ches u p ea ch origin
loca t ion wit h t h e n ea r es t zon e, ea ch d es t in a t ion loca t ion wit h t h e n ea r es t zon e, a n d
calcula t es t h e n u m ber of t r ips from ea ch origin zon e t o each dest in a t ion zon e. Th is is a n
observed dist r ibut ion of t r ips by zon e.
The steps in r un ning the model ar e as follows:
1. Ca lc u l a te o b s e r v e d o r ig i n -d e s t i n a t io n t r ip s . Ch eck if a n em pir ica l

or igin-dest ina t ion t r ip dist r ibut ion is t o be ca lcu lat ed.
2. Origi n file . Th e origin file is a list of origin zon es wit h a sin gle point
r epr esen t ing t h e zon e (e.g., t h e cen t r oid). It m u st be inpu t a s eit h er t h e
pr im a r y or secon da r y file. Specify wh et h er t h e da t a file is t h e pr im a r y or
secon da r y file.
14.15
Figure 14.3:
Trip Distribution Module

O r i g i n ID . S pecify t h e or igin ID va r ia ble in t h e da t a file (e.g., Ce n su sTr a ct ,
Block , TAZ).
3. D e s tin a tio n fi le . Th e dest in a t ion file is a list of dest in a t ion zon es wit h a
sin gle point r epr esen t in g t h e zone (e.g., t h e cent r oid). It m u st be in pu t a s
eit h er t h e pr im a r y or secon da r y file. Sp ecify wh et h er t h e da t a file is t h e
pr im a r y or secon da r y file.Sp ecify t h e dest in a t ion ID va r ia ble in t h e da t a file
(e.g., Cen su sTr a ct , Block , TAZ).
Note: all destinat ion IDs should be in t he origin zone file and m ust ha ve the sam e
na mes an d both should be cha ra cter (string) var iables.
4. Se le ct da ta file . Th e da t a set m u st h a ve in divid u a l or igin a n d dest in a t ion

loca t ion s. E a ch r ecor d m u st h a ve t h e X/Y coor din a t es of a n or igin loca t ion
a n d t h e X/Y coord in a t es of a dest in a t ion locat ion . For exa m ple, an a r r est file
m igh t list in divid u a l in cid en t s wit h ea ch in cid en t h a vin g a cr im e loca t ion
(th e dest in a t ion ) an d a r esiden ce or a r r est locat ion (th e origin ). Select t h e file
th at ha s th e X an d Y coordina tes for t he origin and destina tion locat ions.
Crim eS tat can r ea d ASCI I, d ba se '.dbf', ArcView '.s h p' a n d Ma pIn fo 'da t '
files. Select t h e t a b a n d specify t h e t yp e of file t o be select ed. U se t h e br owse
bu t t on t o sea r ch for t h e file. If t h e file t yp e is ASCI I, s elect t h e t yp e of da t a
sepa r a t or (comm a , sem icolon , sp a ce, ta b) an d t h e n u m ber of colu m n s.
Va r ia bl es. Defin e t h e file wh ich con t a in s t h e X an d Y coordin a t es for bot h

t h e or igin (r esid en ce) a n d dest in a t ion (cr im e) loca t ion s.
C ol u m n . S elect t h e va r ia bles for t h e X an d Y coordin a t es r espect ively for

both t h e or igin a n d dest in a t ion loca t ion s (e.g., Lon , La t , H om eX, H om eY,
In ciden t X, Incident Y.) Bot h loca t ion s m u st be defined for t h e r ou t ine t o
w or k .
Missi n g v a lu es. Identify wheth er th ere ar e any missing values for t hese
four fields (X an d Y coordina tes for both origin and destina tion locat ions). By
defau lt , Crim eS tat will ignore r ecor ds wit h bla n k va lu es in a n y of th e eligible
fields or r ecor ds wit h n on -n u m er ic valu es (e.g.,alph a n u m er ic ch a r a ct er s, #,
*). Bla n ks will a lways be exclud ed u n less t h e u ser select s <n on e>. Th er e
ar e 8 possible options:
1. <bla n k > fields a r e a u t oma t ically exclud ed. Th is is t h e defau lt

2. <n one> in dica t es t h a t n o record s will be exclud ed. If th er e is a
bla n k field, Crim eS tat will tr eat it a s a 0
3. 0 is exclud ed
4. 1 is exclud ed
5. 0 a n d 1 ind ica t es t h a t bot h 0 an d -1 will be exclud ed
6. 0, -1 an d 99 99 in dica t es t h a t a ll t h r ee va lu es (0, -1, 9999) will
be exclud ed
14.17
An y ot h er n u m er ica l va lu e ca n be t r ea t ed a s a m is sin g va lu e by t yp in g it
(e.g., 99)Mu ltiple n u m er ica l va lues ca n be tr ea t ed a s m issin g va lues by
t yp in g t h em , s epa r a t in g ea ch by comm a s (e.g., 0, -1, 99, 9999, -99).
T yp e o f c oo r d i n a t e s ys t em a n d d a t a u n i t s. Th e coor d in a t e s ys t em a n d
dat a u nits a re listed for inform at ion. If th e coordina tes ar e in longitu des and
la t it u des, t h en a sp h er ical syst em is bein g used a n d d a t a u n it s will
a u t oma t ically be decim a l degr ees . If th e coord in a t e syst em is p r oject ed (e.g.,
St a t e P la n e, Un iver sa l Tr a n sver se Mer cat or U TM), th en da t a u n it s cou ld
be eit h er in feet (e.g., St a t e P la n e) or m et er s (e.g., U TM .).
5. Ta ble o u tp u t. Th e fu ll or igin -d es t in a t ion m a t r ix is ou t p u t a s a t a ble t o t h e

screen including summ ar y file inform at ion a nd:
1. The origin zone (ORIGIN)

2. Th e dest in a t ion zon e (DE ST)
3. The n u m ber of obser ved tr ips (FRE Q)
6. S a v e o bs e rv e d o ri g in -de s tin a tio n tri p s . If sp ecified, t h e full or igin-

dest in a t ion m a t r ix ou t pu t is s a ved a s a dbf file n a m ed by t h e u ser . Th e file
out put includes:

2. Th e dest ina t ion zon e (DE ST)
3. Th e X coor din a t e for t h e or igin zon e (ORIGINX)
4. Th e Y coor din a t e for t h e or igin zon e (ORIGINY)
5. Th e X coor din a t e for t h e dest in a t ion zon e (DE STX)
6. Th e Y coor din a t e for t h e dest in a t ion zon e (DE STY)
7. The n u m ber of t r ips (FRE Q)
Note : ea ch r ecor d is a u n ique origin -dest ina t ion com bina t ion . Ther e
a r e M x N r ecor ds wh er e M is t h e n u m ber of origin zon es (in clud in g
th e extern al zone) an d N is the nu mber of destinat ion zones.
7. S a v e l in k s . The top observed origin-destinat ion t rip links can be saved as

sepa r a t e li ne object s for u se in a GIS. Specify t h e ou t pu t file for m a t
(ArcView '.sh p', M apIn fo '.m if' or Atlas*GIS '.bn a ') a n d t h e file n a m e.
S a v e t o p li n k s . Beca u se t h e ou t pu t file is very lar ge (n u m ber of or igin zon es

x n u m ber of dest in a t ion zon es), t h e u ser ca n select a su b-set of zon e
com bina t ion s with t h e most observed t r ips. Indica t ing th e top K link s will
n a r r ow t h e n u m ber down t o t h e m ost im por t a n t on es. Th e defa u lt is t h e t op
100 or igin -dest in a t ion com bin a t ion s. E a ch ou t pu t object is a lin e fr om t h e
or igin zon e t o t h e dest ina t ion zon e with a n ODT pr efix. The pr efix is placed
befor e t h e ou t pu t file n a m e. Th e lin e gr a ph ica l ou t pu t for ea ch object
includes:
14.18
1. An I D n u m b er fr om 1 t o K , w h er e K is t h e n u m b er of lin k s
ou t pu t (ID)
2. Th e fea t u r e pr efix (ODT)
9. The n u m ber of obser ved tr ips for t h a t com bina t ion (FRE Q)
10. Th e dist a n ce bet ween t h e or igin zon e a n d t h e dest ina t ion zon e.
S a v e p oi n t s. In t r a -zon a l t r ip s (t r ip s in wh ich t h e or igin a n d d es t in a t ion a r e

t h e sa m e zon e) ca n be ou t pu t a s sep a r a t e p oi nt objects a s an ArcView '.sh p',
M apIn fo '.m if' or Atlas*GIS '.bn a ' file. Aga in , t h e t op K poin t s a r e ou t pu t
(d efa u lt =100). E a ch ou t pu t object is a p oin t r ep res en t in g a n in t r a -zon a l t r ip
wit h a n ODTP OINTS pr efix. Th e pr efix is pla ced before t h e out pu t file
n a m e.
The point graph ical out put for each object includes:
1. An I D n u m b er fr om 1 t o K , w h er e K is t h e n u m b er of lin k s
ou t pu t (ID)
2. Th e fea t u r e pr efix (P OIN TSODT)
9. The n u m ber of obser ved tr ips for t h a t com bina t ion (FRE Q)
Exam ple of Observed Distribution from Baltimo re Coun ty
F igu r e 14.4 sh ows t h e ou t pu t of t h e t op 1000 lin k s for t h e obs er ved t r ip dis t r ibu t ion
fr om a sa m ple of 41,974 r ecor ds for inciden t s com m itt ed bet ween 1993 an d 1997. The
zon a l model used wa s t h a t of t r a ffic a n a lysis zon es (TAZ). These wer e discuss ed in cha pt er
12. Beca u se t h er e a r e a la r ge n u m ber of lin k s (532 or igin zon es by 325 dest in a t ion zon es),
t h e t op 1000 we r e t a k en . Th es e a ccoun t ed for 19,615 cr im e t r ips (or 46.7% of all t r ips ). A
la r ger n u m ber of lin k s cou ld h a ve been select ed, bu t t h e m a p wou ld h a ve becom e m or e
clu t t er ed. Of t h e 19,615 t r ip s t h a t a r e dis pla yed in t h e m a p, 7,913 or 40.3% a r e in t r a -zon a l
t r ips . Th ese wer e out pu t by t h e r out in e a s p oin t s a n d h a ve been dis pla yed a s cir cles wit h
t h e size pr opor t ion a l t o t h e n u m ber of t r ip s. Th e r em a in in g 11,702 t r ip lin k s wer e ou t pu t
by t h e r ou t in e a s lin es a n d a r e dis pla yed wit h t h e t h ick n ess a n d st r en gt h of color of th e
lin e bein g pr opor t ion a l t o th e n u m ber of t r ips .
14.19
Figure 14.4:
Observed Baltimore County Crime Trips: 1993-1997
Top 1000 Links
All Crime Types
#
# #
Baltimore County #
#
Top 1000 observed trips

#
# #
# 25 or less
#
# #
#
#
#
26 - 49
#
# #
50 - 74
# # #
#
#
#
# #
75 - 99
#
#
#
#
#
# # # #
100 or more
Top 1000 Intra-zonal observed trips
# # #
#
#
# # #
#
# # #
#
# #
#
Less than 50
#
#
# #
#
#
# #
#
#
#
# #
#
# # #
#
#
50 - 99
#
#
# # #
# #
#
#
100-14 9
# #
# # # # # #
#
#
# # # #
# #
# # # #
# # #
# # # # # #
#
#
# # # #
#
# #
# # #
# #
# # # # #
150-19 9
# # # # # #
# # #
#
#
# #
# # # #
# # #
# # #
#
#
# # # # # # # #
#
#
#
##
#
#
# #
# # #
#
#
# #
#
# City of Baltimore # # #
#
#
#
#
#
# #
#
#
#
#
#
#
#
#
#
200 or more
#
# # #
###
#
#
#
#
# #
#
# #
# #
#
##
#
# # #
# #
#
# #
#
# #
#
Baltimore County
#
# #
# #
#
# #
#
# #
#
#
#
City of Baltimore
#
# # # #
#
#
# # #
#
# #
# #
# #
# ##
# #
##
# #
# #
#
#
#
#
#
# #
# #
#
## #
#
# #
#
#
#
# #
#
#
#
#
0 10 20 Miles W E
S
Th er e a r e sever a l cha r a cte r ist ics of t h e t r ip p a t t er n t h a t sh ould be n ote d. F ir st , t h e
in t r a -zon a l t r ip s t en d t o con cen t r a t e on t h e ea s t er n p a r t of Ba lt im or e Cou n t y. Th is is a n
a r ea t h a t is r elat ively poor with a h igh n u m ber of pu blic h ou sin g project s. This s u ggest s
t h a t t h er e a r e a lot of in t r a -com m u n it y cr im es bein g com m it t ed in t h ese loca t ion s. Secon d,
t h e zon e-t o-zon e pa t t er n , on t h e ot h er h a n d, t en ds t o con cen t r a t e a t five differ en t loca t ion s
r ela t ively clos e t o bor der wit h t h e Cit y of Ba lt im or e. Th ese five loca t ion s a r e a ll m a jor
sh oppin g m a lls. Th ir d, t h e or igin s for t h ose t r ip s t o t h e sh oppin g m a ll t en d t o com e fr om
with in th e City of Baltimore. Four th , in general, th e locat ions with h igh int ra -zona l tr ips
do n ot h a ve a la r ge n u m ber of zon e-to-zon e t r ips . H owever , t h er e is on e except ion in t h e
sou t h west cor n er of t h e cou n t y.
In ot h er wor ds, t h e obser ved dist r ibut ion of cr ime t r ips is com plex, but with severa l
pa t t er n s bein g sh own. A lot of crim e t r ips occur over ver y sh ort dis t a n ces. Bu t t h er e is
a lso a con ver gen ce of ma n y crim e t r ips on m a jor sh oppin g m a lls in t h e Cou n t y.
Ca l ib ra t e Im p e d a n c e F u n c t i o n
Th is r ou t in e a llows t h e ca libr a t ion of a n a ppr oxim a t e t r a vel im peda n ce fu n ct ion

ba sed on a ct u a l t r ip dis t r ibu t ion s. It is u sed t o descr ibe t h e t r a vel im peda n ce in dis t a n ce
only of a n a ct u a l s a m ple (t h e ca libr a t ion s a m ple). U n lik e t h e r em a in in g r ou t in es in t h is
section, t h e Ca libr a t e im peda n ce fu n ction can n ot u se t r a vel t im e, or cost . A file is in pu t
wh ich h a s a set of in cid en t s (r ecor ds) t h a t in clu de bot h t h e X an d Y coordin a t es for t h e
loca t ion of t h e offen der 's r esid en ce (or igin ) a n d t h e X an d Y coordin a t es for t h e loca t ion of
t h e in cid en t t h a t t h e offen der comm it t ed (dest in a t ion .)
Th e r ou t ine est ima t es a t r a vel dist a n ce fu n ct ion u sin g a one-dimen sion a l ker n el
den sit y m et h od. See t h e det a ils in ch a pt er 9. E ssen t ia lly, for ea ch r ecor d, t h e sepa r a t ion
bet ween t h e or igin loca t ion a n d t h e dest in a t ion loca t ion is ca lcu la t ed a n d is r epr esen t ed on
a dis t a n ce sca le. Th e m a xim u m im peda n ce is ca lcu la t ed a n d divid ed in t o a n u m ber of
in t er va ls ; t h e defa u lt is 100 equ a l s ized in t er va ls , bu t t h e u ser ca n m odify t h is . F or ea ch
im peda n ce poin t calcu la t ed, a one-dim en sion a l ker n el is overla id. F or ea ch in t er va l, t h e
valu es of a ll ker n els a r e su m m ed t o pr odu ce a sm oot h fu n ct ion of t r a vel imp eda n ce. The
r es u lt s a r e sa ved t o a file t h a t can be u se d for t h e origin -des t in a t ion m odel.
Not e, h owever , t h a t t h is is a n em pir ica l d is t r ibu t ion a n d r epr esen t s t h e com bin a t ion
of or igin s , d es t in a t ion s , a n d cos t s. It is n ot n eces sa r ily a good des cr ip t ion of t h e im p ed a n ce
(cos t ) fu n ct ion by it s elf. Ma n y of t h e m a t h em a t ica l fu n ct ion s pr od u ce a bet t er fit t h a n t h e
em pir ica l im peda n ce fu n ct ion .
Th e st eps in ca lcu la t in g a n em pir ica l im peda n ce fu n ct ion a r e a s follows:
1. S e l e c t d a ta fi le f o r c a li b ra t i on . Select t he file tha t h as t he X an d Y

coordina tes for t he origin and destina tion locat ions. Crim eS tat ca n r ea d
ASCII, dba se '.dbf', Ar cView '.sh p' an d Ma pIn fo 'dat ' files. Select t h e t a b a n d
select t h e t yp e of file t o be select ed. U se t h e br owse bu t t on t o sea r ch for t h e
14.21
file. If t h e file t yp e is ASCII, s elect t h e t yp e of da t a sepa r a t or (com m a ,
semicolon, space, ta b) an d th e num ber of column s.
Va r ia bl es. Defin e t h e file wh ich con t a in s t h e X an d Y coordin a t es for bot h

t h e or igin (r esid en ce) a n d dest in a t ion (cr im e) loca t ion s
C ol u m n s. Select t h e va r ia bles for t h e X an d Y coordin a t es r espect ively for

both t h e or igin a n d dest in a t ion loca t ion s (e.g., Lon , La t , H om eX, H om eY,
In ciden t X, Incident Y.) Bot h loca t ion s m u st be defined for t h e r ou t ine t o
w or k .
Missi n g v a lu es. Identify wheth er th ere ar e any missing values for t hese
four fields (X an d Y coordina tes for both origin and destina tion locat ions). By
defau lt , Crim eS tat will ignore r ecor ds wit h bla n k va lu es in a n y of th e eligible
fields or r ecor ds wit h n on -n u m er ic valu es (e.g.,alph a n u m er ic ch a r a ct er s, #,
*). Bla n ks will a lways be exclud ed u n less t h e u ser select s <n on e>. Th er e
ar e 8 possible options:
1. <bla n k > fields a r e a u t oma t ically exclud ed. Th is is t h e defau lt

2. <n one> in dica t es t h a t n o record s will be exclud ed. If th er e is a
bla n k field, Crim eS tat will tr eat it a s a 0
3. 0 is exclud ed
4. 1 is exclud ed
5. 0 a n d 1 ind ica t es t h a t bot h 0 an d -1 will be exclud ed
6. 0, -1 an d 99 99 in dica t es t h a t a ll t h r ee va lu es (0, -1, 9999) will
be exclud ed
An y ot h er n u m er ica l va lu e ca n be t r ea t ed a s a m is sin g va lu e by t yp in g it
(e.g., 99)Mu ltiple n u m er ica l va lues ca n be tr ea t ed a s m issin g va lues by
t yp in g t h em , s epa r a t in g ea ch by comm a s (e.g., 0, -1, 99, 9999, -99).
T yp e of c oor d i n a t e s ys t em a n d d a t a u n i t s. Select t h e t ype of coor din a t e

sys t em . If t h e coor din a t es a r e in lon git u des a n d la t it u des, t h en a sph er ica l
system is being used a nd da ta un its will au tomat ically be decimal degrees. If
th e coordina te system is projected (e.g., Sta te Plan e, Universal Tran sverse
Mer ca t or UTM), t h en da t a u n its cou ld be eith er in feet (e.g., Sta t e Pla n e)
or m et er s (e.g., UTM.) Dir ection a l coor din a t es a r e n ot a llowed for t h is
r ou t ine.
2. Se lec t Kern e l P ara m e te rs . Th er e a r e five p ar a m et er s t h a t m u st be

defined.
Meth od of in ter p ola ti on . Th er e a r e fives t yp es of ker n el d is t r ibu t ion s t h a t

can be u se d t o est im a t e point den sit y:
14.22
1. Th e n o r m a l ker n el over la ys a t h r ee-dim en sion a l n or m a l
dist r ibut ion over ea ch poin t t h a t t h en exten ds over t h e a r ea
defin ed by th e r efer en ce file. This is t h e defau lt k er n el
fun ction .
2. Th e u n iform ker n el overla ys a u n ifor m fu n ct ion (disk ) over
ea ch poin t t h a t on ly exten ds for a lim ited dist a n ce.
3. Th e qua rtic kern el overlays a qua rt ic fun ction (inverse
sph er e) over ea ch poin t t h a t on ly exten ds for a lim ited
dis t a n ce.
4. Th e t ri a n g u l a r ker n el overla ys a t h r ee-dim en sion a l t r ia n gle
(cone) over ea ch p oint t h a t only ext en ds for a lim it ed dis t a n ce.
5. Th e n e g a t i v e e x p o n e n t i a l ker n el overla ys a t h r ee
dim en sion a l nega t ive expon en t ial fu n ct ion ('salt sh a ker ') over
ea ch poin t t h a t on ly ext en ds for a lim it ed dis t a n ce
Th e m et h ods p r odu ce sim ilar r esu lt s t h ough t h e n orm a l is gen er a lly

smooth er for a ny given bandwidth.
Ch oi ce o f ban dw idth . The k er n els a r e a pplied t o a lim ited sea r ch dist a n ce,
ca lled 'ba n dwid th '. F or t h e n or m a l k er n el, ba n dwid th is th e s ta n da r d
d evia t ion of t h e n or m a l d is t r ibu t ion . F or t h e u n ifor m , qu a r t ic, t r ia n gu la r
a n d n egat ive expon en t ial k er n els, ban dwidt h is t h e r a diu s of a circle defin ed
by th e su r fa ce. For a ll t ypes, lar ger ba n dwidt h will pr odu ce sm oot h er
den sit y est ima t es a n d both a da pt ive a n d fixed ban dwidt h int er vals can be
selected.
1. Fi xed b a n d w id t h . A fixed ba n dwidt h dist a n ce is a fixed

in t er va l for ea ch p oin t . Th e u s er m u s t defin e t h e in t er va l, t h e
int er val size, a n d t h e dist a n ce u n its by which it is calcu lat ed
(miles , na u t ical m iles, feet , kilom et er s, m et er s.) Th e defau lt
ba n dw idt h set t in g is fixed wit h in t er va ls of 0.25 m iles ea ch.
The int erval size can be cha nged.
2. Ad a p t i v e b a n d w i d t h . An a da pt ive ba n dw idt h dis t a n ce is

id en t ified by t h e m in im u m n u m ber of ot h er poin t s fou n d
wit h in a sym m et r ical ba n d d r a wn a r oun d a sin gle point . A
sym m et r ical ba n d is pla ced over ea ch d ist a n ce point , in t u r n ,
a n d t h e wid t h is in crea sed u n t il t h e m in im u m sa m ple size is
r ea ch ed. Thu s, each p oint h a s a differ en t ba n dwidt h size. The
u ser ca n m odify t h e m in im u m sa m ple size. Th e defa u lt for t h e
ada ptive ban dwidth is 100 points.
S p e ci fy In t er p o la t i on Bi n s. Th e in t er pola t ion bin s a r e defin ed in on e of

two ways:
14.23
1. By t h e n u m ber of bin s. Th e m a xim u m dis t a n ce ca lcula t ed is
divided by t h e n u m ber of sp ecified bin s. Th is is t h e defau lt
wit h 100 bin s. Th e u ser can cha n ge t h e n u m ber of bin s.
2. By t h e dist a n ce bet ween bin s. Th e u ser can sp ecify a bin widt h

in miles, na ut ical miles, feet, kilomet ers, and m eters.
3. Output (Areal) Units . Specify th e den sit y un its a s point s per m ile,
n a u t ica l mile, foot , kilom et er , or m et er . The defau lt is point s per m ile.
4. Calcu late De ns ities o r Probab ilities . Th e den sit y est im a t e for ea ch cell
can be ca lcula t ed in one of t h r ee w a ys:
1. Ab so lu te d e n si ti e s. Th is is t h e n u m ber of point s p er gr id cell a n d is

sca led so t h a t t h e su m of a ll gr id cells equ a ls t h e sa m ple size.
2. R e la ti ve d e n si ti e s. For ea ch gr id cell, th is is t h e a bsolu t e den sit y

divid ed by t h e gr id cell a r ea a n d is expr es se d in t h e out pu t u n it s (e.g.,
poin t s per squ a r e m ile)
3. P r ob ab ili ti e s. Th is is t h e pr opor t ion of a ll inciden t s t h a t occur in

t h e gr id cell. Th e su m of a ll gr id cells equ a ls a pr obabilit y of 1.
Un lik e t h e J t c ca libr a t ion r out in e, t h is is t h e defa u lt . In m ost cas es , a
user would want a pr oport iona l (probability) distr ibut ion a s th e
r ela t ive differ en ces in im peda n ce for differ en t cost s a r e wh a t is of
interest.
Select wh et h er a bsolu t e den sit ies, rela t ive den sit ies, or pr oba bilit ies
ar e to be out put for each cell. The defau lt is probabilities.
5. Se le ct Out pu t F ile . The out put m u st be s a ved t o a file. Crim eS tat ca n s a ve

t h e ca libra t ion ou t pu t t o eith er a dba se 'dbf' or ASCII t ext 't xt' file.
6. Ca li bra te ! Click on 'Ca libr a t e!' t o r u n t h e r ou t in e. Th e ou t p u t is sa ved t o t h e

specified file u pon click in g on 'Close'.
7. Gr a p h i n g th e t r a v e l i m p e d a n c e fu n c t i o n . Click on 'View gr a ph ' to see

t h e t r a vel im peda n ce fu n ct ion . Th e scr een view ca n be pr in t ed by click in g on
'P r in t '. F or a bet t er qu a lit y gr a ph , h owever , t h e ou t pu t sh ou ld be im p or t ed
in t o a gr a ph ics or sp r ea ds h eet pr ogra m .
Exam ple of Empirical Impe dan ce from Baltimo re Coun ty
An exa m ple of a n em pir ical im peda n ce fu n ction from Ba lt im ore Coun t y is seen in
figu r e 14.5. Th is wa s der ived fr om t h e 41,974 in cid en t s in wh ich bot h t h e cr im e loca t ion
a n d t h e offen der s or igin loca t ion wer e kn own . As seen , t h e fu n ct ion look s sim ila r t o a
14.24
Figure 14.5:
Empirical Impedance Function:

All Crimes
0.12
0.1
Proportion of All Trips
0.08
0.06
0.04
0.02
0
0 5 10 15 20 25 30 35 40
Distance (miles)
n egat ive expon en t ial fu n ct ion . But t h er e is a lit t le h itch a r ou n d 3 m iles wher e t h e t r a vel
lik elih ood in cr ea ses, r a t h er t h a n decr ea se. Th is cou ld possibly be du e t o t h e Cit y of
Ba lt im or e bor der wh ich a bu t s m u ch of t h e s ou t h er n pa r t of t h e Cou n t y.
Wh a t ever t h e r ea son , t h e em pir ica l im peda n ce fu n ct ion ca n be u sed a s a pr oxy for

t r a vel cost by offend er s. As we s h a ll see, however , it m a y not p r odu ce as good a fit in t h e
gra vit y model as s om e of t h e m a t h em a t ica l fu n ct ion s, pa r t icu lar ly t h e lognorm a l. The
r ea son is t h a t it is a beh a vior a l d escr ip t ion . Con sequ en t ly, t h e pa t t er n r eflect s bot h t h e
exist en ce of cr im e oppor t u n it ies (a t t r a ct ion s) a s well a s cost s. Wh ile a n em pir ica l
descr ip t ion is u sefu l for gu essin g wh er e a ser ia l offen der m igh t live, for a t r ip dis t r ibu t ion
m odel it a pp a r en t ly does n ot clea n ly est im a t e t h e r ea l cost s t o an offend er . Never t h eles s, it
is a t ool th at can be used.
S e tu p o f Ori gi n -D e st in a ti on Mo d e l
Th e pa ge is for t h e set u p of t h e or igin-dest ina t ion m odel. All t h e r eleva n t files,

m odels a n d expon en t s a r e inpu t on t h e pa ge an d it a llows t h e t r ip dist r ibut ion m odel t o be
ca libra t ed a n d a lloca t ed. Figur e 14.6 shows t h e set u p screen . Th er e a r e a n u m ber of
par am eters t ha t h ave to be defined:
1. P re d ic te d o ri g in fi le . Th e pr edicted origin file is a file t h a t list s t h e origin

zon es wit h a sin gle point r epr esen t in g t h e zone (e.g., t h e cent r oid) a n d a lso
inclu des t h e pr edict ed n u m ber of cr imes by or igin zon e. The file m u st be
in pu t a s eit h er t h e pr im a r y or secon da r y file. Sp ecify wh et h er t h e da t a file is
t h e pr ima r y or secon da r y file.
Or ig in va r ia bl e. Sp ecify t h e n a m e of t h e va r ia ble for t h e pr edicted origin s

(e.g., P RE DICTE D, ADJ ORI GINS).
O r i g i n ID . Specify t h e or igin ID va r ia ble in t h e da t a file (e.g., Ce n su sTr a ct ,

Block, TAZ).
2. P re di ct e d d e st in at io n file . Th e pr edict ed dest in a t ion file is a list of

dest in a t ion zon es wit h a sin gle point r epr es en t in g t h e zone (e.g., t h e
cen t r oid ) a n d a ls o in clu des t h e pr edict ed n u m ber of cr im es by d est in a t ion
zon e. It m u st be in pu t a s eit h er t h e pr im a r y or seconda r y file. Specify
wh et h er t h e da t a file is t h e pr ima r y or secon da r y file.
D e st i n a t i o n v a r i a b l e. Specify t h e n a m e of t h e va r ia ble for t h e

pr edict ed dest in a t ion (e.g., P RE DICTE D, ADJ DE ST).
D e s ti n a ti o n ID . Specify t h e dest in a t ion ID va r ia ble in t h e da t a file

(e.g., Cen su sTr a ct , Block , TAZ).
14.26
Figure 14.6:
Trip Distribution Model Setup

Not e: if M is t h e n u m ber of r ows a n d N is t h e n u m ber of colu m n s, t h en t h e t ot a l
n u m ber of grid cells (M x N) ca n n ot be grea t er t h a n SQRT(RAM - 64)/56 wh er e
RAM is t h e a vailable RAM. Ther e is a m a xim u m a llowa ble of 4 Gb.
3. Expon en ts . Th e exponen t s a r e power t er m s for t h e pr edicted origin s a n d

dest in a t ion s. Th ey in dica t e t h e r ela t ive st r en gt h of t h ose va r ia bles. F or
exam ple, com pa r ed t o a n expon en t of 1.0 (t h e defau lt), a n expon en t grea t er
t h a n 1.0 will st r en gth en t h a t var iable (or igins or dest ina t ion s) while an
expon en t less t h a n 1.0 will wea ken t h a t var iable. They ca n be con sider ed
fin e t u n in g a dju st m en t s.
Or igi ns. Specify th e expon en t for t h e pr edict ed origin s. The d efa u lt is 1.0.
D es t i n a t i on s. Sp ecify t h e exponen t for t h e pr edicted origin s. Th e defau lt is

1.0.
4. Im p e d a n c e f u n c t io n . The t r ip dist r ibut ion r ou t ine can u se t wo differ en t

t r a vel d is t a n ce fu n ct ion s:
1) An a lr ea dy-ca libr a t ed dis t a n ce fu n ct ion ; a n d

2) A m a t h em a t ica l for m u la (t h e d efa u lt ).
U se a n a l r ea d y-c a l i b r a t ed d i st a n ce fu n ct i on . I f a t r a vel d is t a n ce
fu n ct ion h a s a lr ea dy been ca libr a t ed (see 'Ca libr a t e im peda n ce fu n ct ion '
a bove), t h e file ca n be dir ect ly in pu t in t o t h e r ou t in e. Th e u ser select s t h e
n a m e of t h e a lr ea dy-calibr a t ed t r a vel dist a n ce fu n ction . Crim eS tat reads
dba se 'dbf', ASCII t ext 't xt', an d ASCII d a t a 'dat ' files.
U se a m a t h em a t i ca l for m u l a . A ma th emat ical form ula can be used

in st ea d of a calibr a t ed dis t a n ce fu n ction. Sim ila r t o th e J our n ey t o cr im e
m odu le (see cha pt er 9), th er e a r e five m a t h em a t ica l fu n ct ion s. They
m ea s u r e a separation bet ween t wo zon es a n d est im a t e a likelih ood va lu e.
Sepa r a t ion can be in t er m s of dis t a n ce, tr a vel t im e, sp eed (which is
con ver t ed in t o tr a vel t im e), or t r a vel cost s.
Ma t h e m a t i ca l fu n ct i on s
Br iefly, th e five fu n ct ion s a r e:
1. Li n e a r. Th e sim plest t ype of dist a n ce model is a linea r fun ction. Th is

m odel p ost u la t es t h a t t h e lik elih ood of tr a velin g t o a zon e from a n oth er by a n
offen der declin es by a con st a n t a m ou n t wit h dis t an ce fr om t h e offen der s
h om e. It is h igh est n ea r t h e offen der s h om e bu t dr ops off by a con st a n t
a m ou n t for ea ch u n it of dist a n ce u n t il it falls t o zero. The for m of t h e lin ea r
equ a t ion is
14.28
f(d ij) = " +$*S ij (14.14)
wh er e f(d ij) is t h e lik elih ood t h a t t h e offen der will t r a vel fr om zon e ii t o a
pa r t icula r loca t ion, j, S ij is t h e separation in dist a n ce, tim e or cost bet ween
t h e offen der s r es ide n ce, I, a n d locat ion j, " is a slop e coefficien t wh ich
defines the fall off in distan ce, an d $ is a const an t. It would be expected th at
t h e coefficien t $ would h a ve a n ega t ive sign sin ce th e lik elih ood s h ould
declin e wit h sepa r a t ion . Th e u ser m u st pr ovide va lu es for A an d $. The
d efa u lt for A is 10 a n d for $ is -1. Wh en t h e fu n ct ion r ea ch es 0 (t h e X axis),
t h e r out in e a u t oma t ically s u bs t it u t es a 0 for t h e fun ction . F igu r e 14.7
illu st r a t es t h is fu n ct ion .
2. N e g a t i v e Ex p on e n t i a l . A sligh t ly m ore com ple x fun ction is t h e n ega t ive

expon en t ia l. In t h is t ype of model, t h e lik elih ood of t r a vel a lso dr ops off with
dista nce. However, the decline is at a const an t rate of declin e, t h u s d r oppin g
qu ickly nea r t h e offen der s h om e u n t il is a ppr oa ch es zer o likelih ood. The
ma th emat ical form of th e negat ive exponent ial is:
-$*S ij
f(d ij) = "*e (14.15)
wh er e f(d ij) is t h e likelih ood t h a t t h e offen der will t r a vel fr om a n or igin zon e,
ii, t o a d es t in a t ion zon e, j, S ij is t h e sepa r a t ion bet ween t h e or igin zon e a n d
t h e des t in a t ion zon e, e is t h e ba se of t h e n a t u r a l logar it h m , " is t h e
coefficient a n d $ is a n expon en t of e . Th e u ser in pu t s va lu es for " - t h e
coefficien t , a n d $ - t h e exp on en t . Th e defa u lt for " is 10 a n d for $ is 1.
Th is fu n ct ion is t h e on e m ost u sed by t r a vel dem a n d m odeler s. It h a s been

r ecom m en ded for u se by t h e F eder a l High wa y Adm inis t r a t ion (NCH RP ,
1995). F igu r e 14.8 illu st r a t es a t yp ica l n ega t ive exp on en t ia l im peda n ce
fu n ct ion .
3. N or m a l. A n or m a l d is t ribu t ion a ss um es th e p ea k lik elih ood is at som e

op tim a l d is t an ce fr om t h e offen der s hom e ba se. Th u s, t h e fu n ct ion r is es to
t h a t dis t a n ce an d t h en declin es. Th e r a t e of in crea se pr ior t o th e opt im a l
dis t a n ce a n d t h e r a t e of decr ea se fr om t h a t dis t a n ce is sym m et r ica l in bot h
dir ection s. Th e m a t h em a t ical for m is:
(S ij - Mean D)
Zij = ------------------- (14.16)
Fd
1 -0.5*Zij2
f(d ij) = " * -------------------- * e (14.17)
Fd * SQRT(2 B)
14.29
Figure 14.7:
Linear Impedance Function
6%
4%
Impedance
2%
0%
0 4 8 12
Distance (miles)
Linear
Figure 14.8:
Negative Exponential Impedance Function
6%
4%
Impedance
2%
0%
0 4 8 12
Distance (miles)
Negative Exponential
wh er e f(d ij) is t h e likelih ood t h a t t h e offen der will com m it a cr ime a t a
pa r t icula r loca t ion, I (defin ed h er e a s t h e cen t er of a gr id cell), S ij is t h e
sepa r a t ion bet ween ea ch origins zone a n d d est in a t ion zon e, Mea n D is t h e
m ea n dis t a n ce inpu t by t h e u se r , Fd is t h e st a n da r d d evia t ion of dist a n ces, e
is th e base of th e nat ur al logar ithm , an d " is a coefficient . The user input s
va lu es for Mea n D, Fd , a n d ". The defau lt va lues a r e 1 for ea ch of t h ese
par am eters.
By ca r efu lly s ca lin g t h e pa r a m et er s of t h e m odel, t h e n or m a l d is t r ibu t ion ca n

be a da pt ed t o a dis t a n ce deca y fu n ct ion wit h a n in cr ea sin g lik elih ood for
n ea r dist a n ces a n d a decrea sin g likelihood for fa r dist a n ces. For exa m ple, by
ch oos in g a s t a nd a r d d evia t ion gr ea t er t h a n t h e m ea n (e.g., Mea n D = 1,S d =
2), t h e dis t r ibu t ion will be sk ewed t o t h e left be ca u se t h e left
ta il of th e norm al distr ibut ion is not evalua ted. Figur e 14.9 illust ra tes a
possible n or m a l im peda n ce fu n ct ion .
4. L og n or m a l . Th e lognorm a l fu n ction is s im ila r t o th e n orm a l except it is

m or e sk ewed, eit h er t o t h e left or t o t h e r igh t . It h a s t h e pot en t ia l of
sh owing a ver y r a pid in crea se n ea r t h e origin wit h a m ore gr a du a l declin e
from a locat ion of pea k like lih ood. Th e m a t h em a t ical for m of t h e fun ction is:
1 -[ ln (S 2 ij)-Mean D ]2 /2 *Fd 2
f(d ij) = "* ---------------------------- *e
S 2 ij * Fd * SQRT(2 B) (14.18)
wh er e f(d ij) is t h e likelih ood t h a t t h e offen der will com m it a cr ime a t a

pa r t icula r loca t ion, I , define d h er e a s t h e cen t er of a gr id cell, S ij is t h e
sepa r a t ion bet ween t h e origin zon e a n d t h e dest in a t ion zon e, Mea n D is t h e
m ea n sepa r a t ion in pu t by t h e u se r , Fd is t h e st a n da r d devia t ion of
se pa r a t ion, e is th e base of th e nat ur al logar ithm , an d " is a coefficien t . Th e
u se r in pu t s M ea n D, Fd , a n d ". The defau lt va lues a r e 1 for ea ch of t h ese
pa r a m et er s. F igu r e 14.10 illu st r a t es a log-n or m a l im peda n ce fu n ct ion t h a t
h a d wide ut ility in severa l stu dies th a t ar e discu ssed below.
5. T r u n c a t e d N e g a t i v e Ex p on e n t i a l . F in a lly, t h e t r u n ca t ed nega t ive

exp on en t ia l is a join ed fu n ct ion m a de u p of t wo dis t in ct m a t h em a t ica l
fu n ct ion s - t h e lin ea r a n d t h e n ega t ive exp on en t ia l. F or t h e n ea r d is t a n ce, a
positive linea r fu n ct ion is defin ed, st a r t ing a t zero likelihood for dist a n ce 0
a n d in cr ea sin g t o d p , a loca t ion of pea k lik elih ood. Th er eu pon , t h e fu n ct ion
follows a n egat ive expon en t ial, declining qu ickly with dist a n ce. The t wo
m a t h em a t ica l fu n ct ion s m a kin g up t h is spline fun ct ion a r e:
Lin ea r : f(d ij) = 0 + $*S ij = $*d ij for S ij $ 0, d i j# d p (14.19)
Nega t ive ->*S ij

E xponen t ia l: f(d ij) = "*e for Xi > S p (14.20)
14.32
Figure 14.9:
Normal Impedance Function
4%
Impedance
2%
0%
0 4 8 12
Distance (miles)
Normal
Figure 14.10:
Lognormal Impedance Function
30%
25%
20%
Impedance
15%
10%
5%
0%
0 4 8 12
Distance (miles)
Figure 14.11:
Truncated Negative Exponential Impedance Function
10%
6%
Impedance
2%
0 4 8 12
-2%
Distance (miles)
Truncated Negative Exponential

wh er e S ij is t h e sepa r a t ion fr om t h e h ome ba se , $ is t h e slope of t h e lin ea r
fu n ct ion (defa u lt =+1) a n d for t h e n ega t ive exp on en t ia l fu n ct ion " is a
coefficient a n d > is an exponent . Since the negative exponent ial only sta rt s
a t a pa r t icula r dis t a n ce, Ma xd ij, ", is a ssu m ed t o be t h e in t er cept if t h e Y-
a xis wer e t r a n sposed t o t h a t dist a n ce. Figur e 14.11 illu st r a t es a t r u n ca t ed
n ega t ive exp on en t ia l im peda n ce fu n ct ion .
Mod e l p a r a m e t er s. F or ea ch m a t h em a t ica l model, t wo or t h r ee differ en t

par am eters m ust be defined:
1. F or t h e n egat ive expon en t ial, th e coefficient a n d expon en t

2. For t he norm al distr ibut ion, the m ean dista nce, sta nda rd
deviat ion a n d coefficient
3. F or logn or m a l d is t r ibu t ion , t h e m ea n d is t a n ce, s t a n da r d
4. For t he linear distr ibut ion, an int ercept an d slope
5. F or t h e t r u n ca t ed n egat ive expon en t ial, a p ea k d ist a n ce, peak
lik elih ood, in t er cept , an d expon en t .
Th e pa r a m et er s will be obt a in ed eit h er fr om a pr eviou s a n a lysis or

fr om a n it er a t ive pr ocess of exper im en t a t ion . See t h e exa m ple below
un der Compa re observed and predicted tr ips.
5. Fine Tun ing Expon en ts . In a ddit ion , for ea ch fu n ct ion , exp on en t s for t h e
a t t r a ct ion a n d pr odu ct ion t er m s ca n a dju st ed. Th is a llows a fin e t u n in g of
t h e im peda n ce fu n ct ion t o be t t er fit t h e em pir ica l d is t r ibu t ion .
6. D i s t a n c e U n i t s . Th e r out in e can calcu la t e im peda n ce in fou r wa ys, by:
1. Dista nce (miles, na ut ical miles, feet, kilomet ers, and m eters)
2. Travel time (minu tes, hour s)
3. Speed (m iles per h ou r , kilom et er s per h ou r ). Speed is t h en con vert ed
in t o t r a vel t im e, in m in u t es .
4. Gen er a l t r a vel cost s (u n specifie d u n it s).
Th ese m u st be set u p u n der Net wor k Dist a n ce on t h e Mea su r em en t

P a r a m et er s p a ge. In t h e N et wor k P a r a m et er s d ia logue, sp ecify t h e
measu rem ent u nits. The defau lt is dista nce in m iles.
7. As su m e d Im p e da n ce fo r E xt e rn a l Zo n e s. For t rips originat ing out side

t h e st u dy ar ea (exter n a l tr ips), specify th e a m ou n t a n d t h e u n its t h a t will be
assu med for t hese trips. The defau lt is 25 miles.
8. As su m e d Im p e da n ce fo r In tra -z on a l Tri ps . F or t r ips origin a t in g a n d

en din g in t h e sa m e zone (in t r a -zon a l t r ips ), sp ecify t h e a m oun t a n d t h e u n it s
th at will be assu med for t hese trips. The defau lt is 0.25 miles.
14.36
9. Mo d e l Co n s tra in ts . In calibr a t in g a m odel, t h e r out in e m u st con st r a in
eit h er t h e or igin s or t h e d es t in a t ion s (s in gle con s t r a in t ) or con s t r a in bot h t h e
origin s a n d t h e dest in a t ion s (dou ble con st r a in t ). In t h e la t t er cas e, it is a n
it er a t ive s olu t ion . Th e d efa u lt is to con st r a in des t in a t ion s a s it is as su m ed
t h a t t h e dest in a t ion t ot a ls (t h e n u m ber of crim es occur r in g in ea ch zon e) a r e
pr obably m ore a ccu r a t e t h a n t h e n u m ber of crim es origin a t in g in ea ch zon e. .
Sp ecify t h e t ype of cons t r a in t for t h e m odel.
Const ra in ori gin s. If con st r a in or igin s is select ed, t h e t ot a l n u m ber of

tr ips from each origin zone will be held const an t.
Con st r a i n d es t i n a t i on s. If con st r a in dest in a t ion s is select ed, t h e t ot a l

n u m ber of t r ips from ea ch d es t in a t ion zon e will be h eld con st a n t .
Con st r a i n b ot h or i g i n s a n d d es t i n a t i on s. I f con s t r a in bot h or igin s a n d

dest ina t ion s is select ed, th e r ou t ine work s out a ba lan ce bet ween t h e
nu mber of origins and t he nu mber of destinat ions.
F i tt i n g th e I m p e d a n c e F u n c t i o n
Th e im peda n ce fu n ct ion is fit in a n it er a t ive m a n n er . F ir st , eit h er a n em pir ica l

im peda n ce or a m a t h em a t ica l im peda n ce is ch osen . Secon d, t h e pa r t icu la r m a t h em a t ica l
fu n ct ion is select ed. F or exa m ple, wit h t h e logn or m a l fu n ct ion , wh ich h a s been fou n d t o
pr odu ce t h e best fit for t h r ee differ en t da t a set s, t h er e a r e t h r ee pa r a m et er s: 1) t h e m ea n
dis t a n ce; 2) th e st a n da r d d evia t ion of dist a n ce; a n d 3) th e coefficien t .
Th ird , in itia l va lues of t h e pa r a m et er s a r e ch osen ; on e su ggest ion is t o u se t h e

defau lts a vailable in t h e Crim eS tat rout ines. The Compa re observed and predicted tr ips
r out in e is u sed t o evalu a t e t h e fit of t h e m odel. F our t h , t h e pa r a m et er s a r e a dju st ed in
sm a ll increm en t s, one a t a t im e, on bot h sid e of t h e in it ia l guess in ord er t o im pr ove t h e fit.
F or exa m ple, wit h t h e lognorm a l fu n ction, t h e m ea n dis t a n ce is fit first becau se it h a s t h e
gr ea t es t im p a ct on t h e over a ll fit . Th en , a ft er a bes t m ea n d is t a n ce h a s been fou n d , t h e
st a n da r d d evia t ion of dist a n ce is a dju st ed u n t il it pr odu ces a bes t fit. Th en , t h e
coefficien t is a dju s t ed u n t il it p r od u ces a bes t fit . F ift h , a n d fin a lly, t h e fin e t u n in g
exp on en t s of t h e p r od u ct ion a n d a t t r a ct ion fu n ct ion s a r e a d ju s t ed . Typ ica lly, t h es e ch a n ge
t h e fina l fit only slight ly. Hen ce, th ey repr esen t a fina l ad just m en t .
Th is pr ocess is illust r a t ed below in t h e discuss ion on t h e com pa r ison of t h e obser ved

a n d pr edict ed t r ips. Es sen t ially, t h e em pir ica l (obser ved) dist r ibut ion is being us ed a s a
calibrat ion sa mple in order t o find tha t impedan ce model an d par am eters t ha t best
a pp roxim a t e it .
Th e O ri g in -D e s t i n a ti o n Mo d e l
Th e t r ip d ist r ibu t ion (or igin-dest in a t ion ) model is im plem en t ed in t wo st eps. F ir st ,

t h e coefficien t s a r e ca lcu la t ed a ccor din g t o t h e exp on en t s a n d im peda n ce fu n ct ion s
14.37
specified on t h e set u p pa ge. Secon d, t h e coefficien t s a n d exp on en t s a r e a pplied t o t h e
pr edict ed origin s a n d des t ina t ion s r esu ltin g in a pr edict ed t r ip dist r ibut ion . Beca u se t h ese
t wo st eps a r e sequ en t ia l, t h ey can n ot be r u n sim u lt a n eously.
Ca li b ra t e Or ig i n -D e s t i n a t io n Mo d e l.
In t h is r ou t ine, t h e r ow or colum n pa r a m et er s (or bot h if double con st r a int is us ed)

a r e est ima t ed u sing a ca libra t ion file. The st eps a r e as follows:
1. Ch e c k t h e Ca libr a t e or igin -dest in a t ion m odel box t o r u n t h e ca libr a t ion

m odel.
2. Sa v e Mod e le d Co e ffic ie n ts (pa ra m e te rs ). The modeled coefficient s ar e

sa ved as a dbf file. Specify a file na m e.
Ap p ly P r e d i c t e d Or ig i n -D e s t i n a t io n Mo d e l
In t h is r ou t ine, t h e coefficient s t h a t wer e ca libra t ed in t h e a bove rout ine can be

a pplied t o a da t a set . The da t a set ca n be th e sa m e a s t h e ca libra t ion file or a differ en t
on e. Th e r ea s on for s ep a r a t in g t h e ca libr a t ion fr om a p plica t ion s t ep s is t h a t t he
coefficient s can be used for m an y different dat a sets. The steps ar e as follows:
1. Ch e c k t h e Apply pr edicted origin -dest in a t ion m odel box to ru n t h e t r ip

dis t r ibu t ion p r ediction .
2. Mode le d Co e fficie n ts Fi le . Loa d t h e m odeled coefficien t s file sa ved in t h e

Ca libra t e or igin-dest ina t ion m odel st a ge.
3. As s u m e d Co o r d in a t e s f or Ex t e r n a l Zo n e . In or der t o m odel t r ip s fr om
t h e ext er n a l zon e (t r ip s fr om ou t sid e t h e st u dy a r ea ), s pecify coor din a t es for
t h is zon e. These coor din a t es will be us ed in d r a wing lines from t h e pr edict ed
or igins t o t h e pr edict ed des t ina t ion s. Ther e a r e fou r ch oices:
1. Mean center (th e mean X an d mean Y of all origin file points a re

t a k en ). Th is is t h e defau lt .
2. Lower -left cor n er (th e m in im u m X a n d m in im u m Y va lu es of a ll or igin
file poin t s a r e t a k en ).
3. Upp er -righ t cor n er (th e m a xim u m X a n d m a xim u m Y va lu es of a ll
or igin file poin t s a r e t a k en ).
4. Use coor din a t es (u ser -defin ed coor din a t es). In dica t e t h e X a n d Y
coordina tes th at ar e to be used.
Beca u s e a n a r bit r a r y loca t ion is t a ken t o r ep r es en t t h e ext er n a l zon e, a n y lin es

t h a t a r e sh own from t h a t zone will n ot n ecessa r ily r epr es en t a n y r ea l t r a vel beh a vior.
H owever, if a very h igh pr oport ion of a ll cr ime t r ips fall wit h in t h e m odeled or igin zon es
14.38
(i.e., 95% or m or e), t h en it is ver y u n lik ely t h a t a n y of t h e t op t r ip lin k s will com e fr om t h e
ext er n a l zone.
4. Ta ble Ou tp u t. Th e t a ble ou t pu t in clu des su m m a r y file in for m a t ion a n d:

3. The num ber of predicted tr ips (PREDTRIPS)
5. S a v e P r e d i c te d O ri g i n -d e s t i n a t io n Tr ip s . Define t h e ou t pu t file. The

out pu t is s a ved a s a dbf file sp ecified by t h e u se r .
6. F ile Ou tp u t. The file out put includes:

7. The num ber of predicted tr ips (PREDTRIPS)
Note : ea ch r ecor d is a u n iqu e or igin -d es t in a t ion com bin a t ion a n d t h er e a r e

M x N r ecor ds wh er e M is th e n u m ber of or igin zon es (in clud ing t h e exter n a l
zon e) a n d N is t h e n u m ber of dest ina t ion zon es.
7. S a v e Li n k s . The top predicted origin-destinat ion t rip links can be saved as

sepa r a t e li ne object s for u se in a GIS. Specify t h e ou t pu t file for m a t
(ArcView '.sh p', M apIn fo '.m if' or Atlas*GIS '.bn a ') a n d t h e file n a m e.
S a v e To p Li n k s
Beca u se t h e ou t pu t file is ver y la r ge (n u m ber of or igin zon es x n u m ber of

dest in a t ion zon es), t h e u ser ca n select a su b-set of zon e com bin a t ion s wit h t h e m ost
pr edict ed t r ip s. In dica t in g t h e t op K lin k s will n a r r ow t h e n u m ber down t o t h e m ost
im por t a n t on es. Th e defa u lt is t h e t op 100 or igin -dest in a t ion com bin a t ion s. E a ch ou t pu t
object is a line from t h e origin zon e t o th e dest in a t ion zon e wit h a n ODT pr efix. Th e pr efix
is p la ced be for e t h e out pu t file na m e.
The gra phical out put includes:
1. An ID n u m ber fr om 1 t o K, wh er e K is t h e n u m ber of lin k s ou t p u t

(ID)
2. Th e fea t u r e pr efix (ODT)
14.39
9. The num ber of predicted tr ips for t ha t combinat ion (PREDTRIPS)
10. Th e dist a n ce bet ween t h e or igin zon e a n d t h e dest ina t ion zon e.
8. Save P oints
In t r a -zon a l t r ip s (t r ip s in wh ich t h e or igin a n d dest in a t ion a r e t h e sa m e zon e) ca n

be ou t pu t a s s ep ar a t e p oi nt objects a s an ArcView '.sh p', M apIn fo '.m if' or Atlas*GIS '.bn a '
file. Aga in , t h e t op K poin t s a r e out pu t (defau lt =100). Ea ch out pu t object is a point
r epr esen t in g a n in t r a -zon a l t r ip wit h a n ODTP OINTS pr efix. Th e pr efix is pla ced befor e
t h e out pu t file na m e.
The gra phical out put for each includes:
1. An ID n u m ber fr om 1 t o K, wh er e K is t h e n u m ber of lin k s ou t p u t

(ID)
2. Th e fea t u r e pr efix (P OIN TSODT)
9. The num ber of predicted tr ips for t ha t combinat ion (PREDTRIPS)
Exam ple of the P redicte d Trip Distributio n from B altimore County
Th e pr edicted origin s a n d p r edicted dest in a t ion s fr om Ba lt im ore Coun t y wer e in pu t

in t o a t r ip dis t r ibu t ion m odel a n d a pr edict ed t r ip dis t r ibu t ion wa s ou t pu t . Th e im peda n ce
fu n ct ion wa s a logn or m a l d is t ribu t ion , wh ich pr od uced a good fit t o t h e obs er ved
(em pir ica l) dis t r ibu t ion (see dis cu ssion be low).
F igu r e 14.12 ou t pu t s t h e t op 1000 lin k s fr om t h e m odel. Th e t op 1000 lin k s a ccou n t

for 14,271.9 t r ips, or 34.0% of t h e t ot a l nu m ber of t r ips. Com pa r ed t o t h e obser ved
dis t r ibu t ion, t h e t op 1000 lin k s a ccoun t for a sm a ller pr oport ion of th e t ota l t r ips (14,272 v.
19,615). This su ggest s t h a t t h e a ct u a l dist r ibut ion is sligh t ly m or e con cen t r a t ed t h a n t h e
m odel su ggest s. Lik e t h e obser ved d ist r ibu t ion , however , a sizea ble n u m ber of t h e t op link s
a r e int r a -zon a l tr ips (5,428 or 12.9%). The int r a -zon a l tr ips h a ve been disp layed a s cir cles
in t h e figu r e.
Com pa r in g t h e pr edict ed t r ip dis t r ibu t ion t o t h e obs er ved t r ip dis t r ibu t ion , s om e
sim ila r it ies a n d differ en ces a r e seen . F igu r e 14.13 com pa r es t h e t op 1000 zon e-t o-zon e
link s for t h e pr edict ed a n d observed dist r ibut ion s. The m odel ha s capt u r ed m a n y of t h e
14.40
Figure 14.12:
Predicted Baltimore County Crime Trips: 1993-1997
Top 1000 Links
All Crime Types
#
# #
Baltimore County #
#
Top 1000 predicted trips

#
# #
# 25 or less
#
# # #
#
#
#
26 - 49
#
# #
50 - 74
# # #
# #
#
#
#
#
75 - 99
#
#
#
#
#
# # #
#
#
100 or more
Top 1000 intra-zonal predicted trips
# # #
#
# # #
# #
# # #
#
#
#
# #
#
# #
#
#
# #
#
Less than 50
#
#
# #
# #
#
# # #
#
#
#
50 - 99
#
#
# #
#
# #
# #
# #
100-14 9
# #
# # # # #
#
#
# # # #
# #
#
# # # # #
# # # #
# # # # # #
#
#
# # # # #
# #
# # #
#
#
#
# # # # # #
150-19 9
# # # # # #
# # #
#
# # #
# # # #
# #
#
# # # # # # #
# # # #
#
#
#
#
# #
# # #
# #
# #
City of Baltimore # # #
#
#
#
# # #
#
#
#
# # #
#
#
# # #
#
# #
# #
#
#
#
# #
#
#
#
200 or more
#
#
#
# #
# #
#
# #
# # #
# #
#
#
# # # #
# # # #
#
# #
# #
# #
#
#
Baltimore County
#
# #
# #
#
# #
# #
#
#
#
#
City of Baltimore
#
#
# # #
# #
# #
# #
# #
#
# #
# # #
# #
#
#
# #
#
#
# # #
#
# # # #
# #
# #
#
# #
##
#
# # #
#
# #
#
# #
#
#
0 10 20 Miles W E
S
Figure 14.13:
Comparison of Predicted and Observed Crime Trips
1000 Top Zone-to-Zone Trips
All Crime Types
Baltimore County
Top 1000 predicted trips

25 or less
26 - 49
50 - 74
75 - 99
100 or more
Top 1000 observed trips
City of Baltimore 25 or less
26 - 49
50 - 74
75 - 99
100 or more
Baltimore County
City of Baltimore
N
0 10 20 Miles
W E
S
m a jor lin ks . For t h e five sh oppin g ma lls th a t r eceived m a n y actu a l cr ime t r ips, th e m odel
h a s ca pt u r ed t h e m a jor it y of t r ip s for t h r ee of t h em a n d som e t r ip s for a fou r t h . F or t h e
ma ll in t he sout heast corn er of th e coun ty, on t he oth er ha nd, the m odel ha s not a llocat ed a
lar ge nu m ber of t r ips. Simila r ly, for a zon e n ea r t h e west er n edge of t h e cou n t y, t h e m odel
ha s allocat ed more tr ips th an actu ally occur red.
Th er e a r e, of cou r se , only 32 5 in t r a -zon a l t r ip lin k s (one for ea ch d es t in a t ion zon e).

Lookin g at a com pa r ison of t h e int r a -zon a l tr ips (figu r e 14.14), som e sim ilar ities a n d
differe n ces a r e seen . Gen er a lly, th e m odel cap t u r ed t h e loca t ion of m a n y int r a -zon a l t r ips ,
bu t it did n ot ca pt u r e t h e qu a n t it y ver y a ccu r a t ely. Zones t h a t h a d m a n y int r a -zon a l t r ips
a r e sh own a s h a vin g on ly som e by t h e m odel a n d, con ver sely, t h e m odel p r edict s m a n y
in t r a -zon a l t r ip s for t wo zon es wh ich h a d on ly s om e.
In oth er wor ds , t h e fit bet ween t h e a ctu a l dis t r ibu t ion a n d t h e m odel is n ot p er fect .
Con sid er in g t h a t on ly 1000 of t h e 172,900 t r ip lin ks (532 or igin zon es x 325 dest in a t ion
zon es) ar e sh own, t h e m odel h a s s t ill done a r ea son a ble job of cap t u r in g t h e m a jor link s
It is n ot s u r pr isin g t h a t t h e m odel is n ot p er fect . Th e m odel is a sim ple a n a logue

u sin g on ly t h r ee va r ia bles (pr odu ct ion s, a t t r a ct ion s, im peda n ce) wh er ea s t h e a ct u a l
dis t r ibu t ion r epr esen t s a ver y comp lex set of in dividu a l decision s m a de by offend er s. Wh a t
is per h a ps r em a r ka ble is t h a t t h e m odel ha s done a decent job of ca pt u r ing some of t h ese
r ela t ion sh ips a t a ll.
Th is br in gs u p a n im por t a n t poin t , n a m ely t h a t a m odel is n ot r ea lit y; it is only a

sim plified set of r elat ion sh ips t h a t a ppr oxim a t es r ea lity (in t h is ca se, th e obser ved
dist r ibut ion ). It is im port a n t in developin g an y model to eva lua t e it r elat ive t o a n obser ved
set of facts, and t his applies no less to th e trip distr ibut ion m odel. One has t o un dersta nd,
h owever, t h a t a good m odel will n ot ca pt u r e a ll t h e r elat ion sh ips. Hopefu lly, it ca pt u r es
en ou gh of t h em t o m a k e t h e m odel u sefu l for pr edict ion a n d eva lu a t in g p olicy op t ion s.
Co m p a r in g Ob s e rv e d & P r e d i c t e d Tr ip s
It is im por t a n t t o cond u ct a n u m ber of t est s on t h e pr edicted m odel t o ens u r e t h a t it

is ca pt u r ing t h e m ost imp or t a n t elemen t s of t h e obser ved dist r ibut ion . These a r e con du ct ed
by com pa r in g t h e pr edicte d d ist r ibu t ion wit h t h e obser ved (em pir ical) dist r ibu t ion.
Th er e a r e a n u m ber of t es t s t h a t can be u se d t o evalu a t e a m odel by comp a r in g t h e

pr edicte d d ist r ibu t ion wit h t h e obse r ved one. Crim eS tat includes thr ee of th ese an d th e
steps ar e as follows:
1. Est imat e the par am eters of th e model an d apply th em to the calibrat ion da ta
set
2. E xam ine t h e int r a -zon a l tr ips t o be su r e t h a t t h e pr edict ed n u m ber

cor r es pon ds to t h e obs er ved nu m ber
14.43
Figure 14.14:

Top Intra-zonalTrips
#
All Crime Types
#
# #
Top intra-zonal trips

#
#
#
Less than 50
#
Baltimore County # 50 - 99
#
#
#
100-149
#
#
#
# #
#
150-199
#
# #
# #
# #
#
# #
# #
# # #
# #
#
#
#
# #
# #
#
#
200 or more
# #
# #
#
#
# # #
# #
#
# #
# #
# # #
Top intra-zonal observed trips

#
#
#
# #
#
# #
#
Less than 50
#
# # # # #
#
#
# # # #
# # #
#
#
50 - 99
# # #
#
# # #
# # #
# # #
#
#
# # # #
# #
# #
# # #
100-149
# # #
# # # #
# ## # #
# #
#
# # # # #
# # # # #
#
# # #
# # # #
#
# #
# # # # #
# # # # # # #
# # #
#
#
# # #
# # # # #
# #
# #
# # #
#
150-199
#
# # # # # # # #
#
#
##
#
# #
#
# # #
#
# #
# #
#
#
#
# # # #
## #
#
#
##
#
#
# #
#
# #
#
# # # #
#
##
# # # #
# # #
# # #
200 or more
#
#
#
#
#
# # ##
#
# # #
#
# #
##
# # #
#
##
#
# # #
# #
# # #
# # # #
#
# # #
# # #
#
# #
Baltimore County
# #
# #
#
# #
# # #
#
City of Baltimore
#
#
# # # #
#
#
# # # # #
# #
# #
# #
###
# #
##
# # # # #
# #
#
# #
#
#
#
#
#
# #
#
## # #
#
#
# #
# #
#
#
# #
#
#
# # # # #
# #
#
#
#
# #
N
0 10 20 Miles
W E
S
3. Com p a r e t h e t r ip len gt h s of t h e obs er ved a n d p r ed ict ed dis t r ibu t ion s u sin g
two tests:
A. Th e Coinciden ce Rat io
B. The Komolgorov-Smirn ov Two-sam ple Test
4. Com p ar e t h e n u m ber of t r ip s for t h e t op lin k s u sin g a ps eu do-Ch i s qu a r e

t est . Tha t is, th e n u m ber of t r ips for t h e m ost fr equ en t lin ks in t h e obser ved
dist r ibut ion a r e com pa r ed t o t h e n u m ber pr edict ed by th e m odel for t h e sa m e
link s.
Un for t u n a t ely, not one of t h ese t est s is su fficient t o valida t e a m odel. Fu r t h er ,

m in im izing t h e discr epa n cy for only one of t h em m a y dist ort t h e oth er s. It is ver y un likely
t h a t t h er e will be a m odel t h a t m in im izes t h e er r ors for a ll t h r ee t es t s. Cons equ en t ly, t h e
u ser will h a ve t o choose a m odel t h a t ba la n ces t h ese fa ct or s in a desir a ble wa y (a n
op tim u m m odel).
E s t im a t i n g Im p e d a n c e P a r a m e t e r s a n d E xp o n e n t s o f Gr a v it y Mo d e l
Wh ile t h is is n ot s t r ictly a n eva lu a t ion t es t , t h is s t ep is ess en t ia l in est im a t in g t h e

par ticular impedan ce para met ers th at ar e used in t he first place. Typically, an an alyst
will a ppr oxim a t e a n imp eda n ce fu n ct ion . Usin g a com pa r ison bet ween t h e obser ved an d
pr edict ed m odels, th e pa r a m et er s can be ad just ed t o pr odu ce a bet t er fit. Th e st eps a r e a s
follows:
1. Th e m odel is est im a t ed wit h a ca libr a t ion da t a set . Th er e is a file of

pr edict ed origin s a n d a n ot h er file of pr edict ed des t ina t ion s; typica lly, th ese
a r e defin ed a s t h e pr ima r y an d secon da r y files res pectively, t h ou gh t h e or der
cou ld be r ever sed or t h e sa m e file us ed for both origin s a n d d est in a t ion s (if
t h e n u m ber of or igin s zon es wa s id en t ica l t o t h e n u m ber of dest in a t ion
zon es).
2. On t h e t r ip dis t r ibu t ion set u p pa ge, s elect t h e t yp e of im peda n ce fu n ct ion

t h a t is t o be u sed, a lr ea dy-ca libr a t ed (em pir ica l) or m a t h em a t ica l. F or t h e
jou r n ey to cr ime r ou t ine, gener a lly t h e em pir ica l fu n ct ion led to bett er
r es u lt s t h a n t h e m a t h em a t ical. H owever , wit h a t r ip d ist r ibu t ion fu n ction , a
m a t h em a t ical fun ction m a y be a s good, if not bet t er . Th is wa s t est ed wit h
t h r ee d a t a set s for Ba lt im ore Coun t y, La s Vega s, a n d Ch icago an d, in a ll
ca s es , a m a t h em a t ica l fu n ct ion (t h e logn or m a l) ga ve a m u ch bet t er fit t h a n
a n em pir ica lly-der ived fu n ct ion (see ch a pt er 17).
3. If a m a t h em a t ica l fu n ct ion is t o be us ed, select t h e t ype of dist r ibut ion . The

defau lt va lu e is a logn orm a l, bu t t h e u ser can choose a n ega t ive expon en t ia l,
a n or m a l, a lin ea r , or a t r u n ca t ed n ega t ive exp on en t ia l fu n ct ion .
14.45
4. F or t h e pa r t icu la r m a t h em a t ica l fu n ct ion , s elect in it ia l gu esses for t h e
pa r a m et er s. For ea ch m a t h em a t ica l model, t wo or t h r ee differ en t
par am eters m ust be defined:
1. F or t h e n egat ive expon en t ial, th e coefficient a n d expon en t

2. For t he norm al distr ibut ion, the m ean dista nce, sta nda rd
3. F or logn or m a l d is t r ibu t ion , t h e m ea n d is t a n ce, s t a n da r d
4. For t he linear distr ibut ion, an int ercept an d slope
5. F or t h e t r u n ca t ed n egat ive expon en t ial, a p ea k d ist a n ce, peak
lik elih ood, in t er cept , an d expon en t .
5. In a ddit ion , th er e a r e expon en t s of t h e pr odu ct ion a n d a t t r a ct ion side t h a t

can be m a de t o fine t u n e t h e m odel. In gen er a l, th ese exponen t s will on ly
a ffect t h e r esu lts sligh t ly, com pa r ed t o t h e bas ic ch oices of t h e t ype of m odel
an d th e selection of values for t he ma in para met ers.
6. Ca libr a t e a n d a pp ly t h e m od el t o t h e ca libr a t ion da t a set . E xa m in e t h e t h r ee

cr it er ia dis cu ssed below t o m in im ize t h e er r or bet ween t h e a ct u a l
dis t r ibu t ion a n d t h a t pr edicte d by t h e m odel.
7. Modify t h e pa r a m et er va lu es sligh t ly.
8. Rep ea t st eps 4 t h r ough 7 u n t il a good fit is fou n d be t ween t h e a ctu a l a n d

pr edicte d d ist r ibu t ion a n d in wh ich t h e er r ors a r e m in im ized a n d opt im ized.
The pr ocess by which t h is is don e is discu ssed below.
Co m p a r i n g In t r a -z o n a l Tr ip s
Th e firs t eva lu a t ion t est is t o comp a r e t h e per cent a ge of t r ips t h a t occur wit h in t h e
sa m e zon e - int r a -zon a l tr ips. The Tr a vel Model Im pr ovemen t P r ogra m m a n u a l in dica t es
t h a t in t r a -zon a l t r ip s sh ou ld r epr esen t t yp ica lly n o m or e t h a n 5% of a ll t r ip s for h om e-t o-
wor k t r ip s; t h a t is , com m u t in g t r ip s (F H WA, 1997, ch a pt er 4). H owever , given t h a t m ost
cr im e t r ip s a r e qu it e sh or t , t h e pr opor t ion of t r ip s t h a t a r e in t r a -zon a l is lia ble t o be m u ch
h igh er . In Ba lt im ore Coun t y, for exa m ple , 19.7% of a ll crim e t r ips wer e in t r a -zon a l.
Id ea lly, th e pr edicted m odel sh ould a lso ha ve 19.7% of all crim e t r ips bein g int r a -zon a l.
Th e Com pa r e obser ved an d pr edict ed t r ip lengt h s r ou t ine is discus sed below. The
r ou t ine out pu t s t h e n u m ber of t r ips t h a t a r e int r a -zon a l in bot h t h e obser ved an d pr edict ed
dis t r ibu t ion s. A good m odel s h ou ld pr odu ce a ppr oxim a t ely t h e sa m e n u m ber of in t r a -zon a l
t r ips in t h e pr edict ed dist r ibut ion a s wh a t a ct u a lly occu r r ed.
14.46
Illu st r a ti on
F or exam ple, in t h e Balt imore Coun t y model displayed in figu r e 14.12 a bove, t h er e
wer e 8,272 int r a -zon a l tr ips in t h e a ct u a l dist r ibut ion (ou t of 41,979). On t h e ot h er h a n d,
t h er e wer e on ly 5,428 in t r a -zon a l tr ips in t h e m odel. In oth er wor ds, t h e pr edict ed m odel
a ss ign ed fewer in t r a -zon a l t r ips t h a n a ctu a lly occur r ed.
It m a y be n ecessa r y t o m odify t h e m odel t o pr odu ce a clos er fit for t h e in t r a -zon a l

t r ips . A sim ple wa y t o do th is t o incr ea se or d ecr ea se t h e r ela t ive im peda n ce in t h e m odel.
So, t o u se t h e exam ple, if t h e pr edict ed m odel is as signin g too few int r a -zon a l tr ips, th en
t h e cost fun ction can be s t r en gt h en ed (i.e., m a k in g t r a vel m ore e xpen sive). In t h is ca se, in
t h e origin a l m odel t h e lognorm a l fun ction wa s u se d w it h a m ea n dis t a n ce of 6.18 m iles . If
t h e m ea n dis t a n ce of t h e im peda n ce fu n ct ion is r edu ced t o 3.5, t h en t h e n u m ber of
pr edicted in t r a -zon a l t r ips in crea ses t o 8,275, alm ost t h e sa m e n u m ber a s occur r ed in t h e
obser ved dis t r ibu t ion .
In oth er wor ds , by de cre a sin g t h e m ea n dis t a n ce for t h e lognorm a l fun ction , t h e

imped a n ce fu n ct ion was st r engt h ened (i.e., m a de m or e expen sive) an d a bett er fit was
creat ed between t he observed and predicted distributions.
In a n d of it self, a m ism a t ch for in t r a -zon a l t r ips bet ween t h e pr edicted m odel a n d

wh a t a ctu a lly occur r ed does n t n ecessa r ily r equ ir e a m odifica t ion of th e gr a vit y fun ction .
Ot h er crit er ia m u st be cons ide r ed, n a m ely h ow well t h e pr edicte d m odel fit s t h e t r ip len gt h
dis t r ibu t ion a n d h ow well t h e pr edict ed m odels ca pt u r es t h e m ost fr equ en t in t er -zon a l
(zon e-t o-zon e) t r ip lin ks . Lat er in t h e discuss ion , th e issu e of opt imizing a m odel by
ba la n cing t h es e differ en t crit er ia will be descr ibed .
Co m p a r e Tr ip Le n g t h D i s t r ib u t i o n
Th e secon d evalu a t ion t est in com pa r ing t h e obser ved with t h e pr edict ed
dist r ibut ion is a calculat ion of t h e t r ip lengt h dist r ibut ion (see st eps below). Beca u se t h e
t r ip d ist r ibu t ion m a t r ix will t ypically be ver y lar ge, m ost cell va lu es will be zer o. Ra r ely
will t h er e be en ough da t a t o cover a ll t h e cells a n d, even if t h er e wa s, t h e sk ewn ess in
cr im e dis t r ibu t ion s will lea ve m ost cells wit h n o da t a . F or exa m ple, for t h e Ba lt im or e
Coun t y m odel, wit h 532 or igin zon es a n d 32 5 de st in a t ion zon es, t h er e will be 172,900 cells
(325 x 532). The calibr a t ion da t a set h a d on ly 41,974 ca ses. Th u s, t h e n u m ber of cells is
m ore t h a n fou r t im es t h e sa m ple size a n d it is n ot p ossible t o fill a ll cells wit h a n u m ber .
Con sequ en t ly, beca u se of t h e la r ge n u m ber of cells wit h zer o cou n t s, on e ca n n ot u se

t h e Chi squ a r e t est t o com pa r e t h e obser ved an d pr edict ed dist r ibut ion s. The Ch i squa r e
t est a ssu m es t h a t , firs t , th e dist r ibut ion is r elat ively n or m a l (wh ich it is n ot sin ce t h e da t a
a r e h igh ly sk ewed) an d, secon d, th a t t h er e a r e a t leas t 5 ca ses p er cell. The la t t er
con dit ion is im possib le given t h e la r ge n u m ber of cells.
Th er efor e, wha t is us u a lly done is t o com pa r e t h e trip length d is t r ibu t ion of t h e

obser ved a n d pr edict ed m odels . Tr ip len gt h is t h e len gt h in dis t a n ce, t r a vel t im e, or cost
14.47
of ea ch t r ip. It is m ea su r ed by th e a ct u a l len gth (or sepa r a t ion ) bet ween t wo zon es t imes
th e num ber of cases for t ha t zone pair. For example, in figur e 14.1, th ere were 15 tr ips
fr om zon e 1 to zon e 2 an d 7 t r ips in t h e opposite d irection (fr om zon e 2 to zon e 1). Lets
assu me th at th e dista nce between zone 1 and zone 2 is 1.5 miles. Thus, ther e are 22 trips
t h a t fall in t o a t r ip len gt h of 1.5 m iles (15 in t h e dir ection of zon e 1 t o zone 2 a n d 7 in t h e
dir ect ion of zon e 2 t o zon e 1).
If t r a vel t ime is u sed, t h e ca lcu lat ion s u ses t ime r a t h er t h a n dist a n ce. For exam ple,
if a vehicle was t r a veling 30 m iles per h ou r , th en it would t a ke 3 m inu t es t o cover 1.5 m iles
(1.5 miles 30 miles per h our = 0.05 hour s x 60 minu tes per hour = 3 minu tes). Thus,
t h er e a r e 22 t r ips t h a t fall in t o a t r ip len gt h of 3 m in u t es . A sim ila r logic would a pp ly t o
tr avel cost cat egories.
Th is pr ocess is r epea t ed for a ll cells a n d t h e dis t r ibu t ion of t r ip s is a lloca t ed t o t h e

dist r ibut ion of t r ip lengt h s (in dist a n ce, tr a vel t ime, or t r a vel cost ). In gen er a l, on e u ses
ma ny int ervals (or bins) for t rip length (25 or m ore). In Crim eS tat, t h e d efa u lt n u m ber of
t r ip len gt h s is 25, bu t it is n ot u n k n own t o us e u p t o 100. Th e pr oblem in u sin g t oo m a n y is
t h a t t h e dist r ibu t ions become u n r elia ble a n d d iffer en ces t h a t a pp ea r m a y n ot be r ea l.
G r a p h i c a l f it
On ce t h e t r ip lengt h dist r ibut ion is ca lcu lat ed for bot h t h e obser ved an d pr edict ed
dis t r ibu t ions , it is p ossible t o comp a r e t h em . Crim eS tat ou t pu t s a gr a ph sh owin g t h e fit of
t h e t wo dist r ibut ion s. In gener a l, t h ey should be ver y close. An exam ina t ion of differ en ces
bet ween t h e dist r ibu t ion s ca n in dica t e a t wh a t t r ip len gt h s t h e m odel is fa iling. This
m igh t a llow t h e pa r a m et er s t o be a dju st ed in ord er t o imp r ove th e fit on t h e n ext it er a t ion.
E xa m ples will be given below of t h e gr a ph in g of t h e t wo dis t r ibu t ion s. Bu t , it s im por t a n t
t o come u p wit h a m odel in wh ich t h e t wo dist r ibu t ion s look sim ila r .
C oi n c i d e n c e r a t i o
Th e coincidence ratio com pa r es t h e t wo t r ip len gt h dis t r ibu t ions by exa m in in g t h e

r a t io of t h e t ot a l a r ea of t h ose dis t r ibu t ion s t h a t coin cid e (i.e., t h a t a r e in com m on ; F H WA,
1997, cha pter 4). It is defined as:
K fO fP
Coin ciden ce = 3 m in [-------, ---------] (14.13)
k=1 FO FP
K fO fP
Tota l = 3 m a x[-------, ---------] (14.14)
k=1 FO FP
14.48
Coin cid en ce
Coin cid en ce r a t io = ------------------- (14.15)
Tot a l
The steps are a s follows:
1. E ssen t ially, t h e t wo dist r ibut ion s a r e br oken int o K bins (or int er vals). Tha t
is, t h e n u m ber of t r ips in ea ch bin is en u m er a t ed (see exa m ple a bove).
2. E a ch of t h e t wo dist r ibu t ion is con ver t ed int o a pr opor t ion by dividin g t h e bin
cou n t by t h e t ot a l n u m ber of t r ip s in t h e d is t r ibu t ion . Th is st ep is n ot
a bs olut ely es se n t ia l a s t h e t es t can be cond u cte d of th e r a w cou n t s.
H owever , by con ver t in g in t o pr opor t ion s, t h e t wo dis t r ibu t ion s a r e
st a n da r dized.
3. A cu m u la t ive cou n t is con d u ct ed of t h e m in im u m pr oport ion in ea ch in t er va l.

Th a t is, st a r t in g a t t h e lowest in t er va l, th e sm a ller of t h e t wo pr opor t ion s is
t a ken . At t h e n ext in t er val, th e sm a ller of t h e t wo pr oport ion s is a dded t o
t h e cou n t . This is r epea t ed for a ll K bins . This is ca lled th e coin cid en ce a n d
measu re th e overlapping proport ions over all int ervals.
4. A s im ila r cu m u la t ive cou n t is con d u ct ed of t h e m a xim u m p r op or t ion in ea ch

int erva l. Tha t is, st a r t ing at t h e lowest int erva l, t h e lar ger of t h e two
pr opor t ion s is t a k en . At t h e n ext in t er va l, t h e la r ger of t h e t wo pr opor t ion s
is a dded t o t h e cou n t . This is r epea t ed for a ll K bins . This is ca lled th e total
an d measu res th e unique proport ion over all int ervals.
5. F in a lly, t h e coin cid en ce r a t io is defin ed a s t h e r a t io of t h e m in im u m cou n t t o

t h e t ot a l cou n t .
Th e coinciden ce r a t io is a p r oport ion fr om 0 to 1. It is an a logou s t o t h e R 2 st a t ist ic

in r egr es sion a n a lysis in t h a t it m ea su r es t h e expla in ed (or overla pp in g) var ia n ce.
Accor din g to th e Tra vel Model Im pr ovemen t P r ogra m m a n u a l (F H WA, 1997, ch a pt er 4),
t h e h igh er t h e coinciden ce r a t io, th e bet t er . A valu e of 0.9 wou ld gener a lly be con sider ed
good .
Kom olgor ov-Sm ir nov tw o-sa m p le test
Th e Kom olgor ov-Sm ir n ov Two-Sa m ple Tes t is s im ila r t o th e coin ciden ce ra t io, bu t it
exa m in es t h e m a xim u m differ en ce across a ll bin s (Ka n ji, 1993). For ea ch d ist r ibu t ion, a
cum u la t ive su m is cr ea t ed. At ea ch in t er va l, t h e differ en ce bet ween t h e t wo cu m u la t ive
su m s is calcu lat ed. The m a xim u m differ en ce bet ween t h e t wo dist r ibut ion s is t a ken a s t h e
t est st a t ist ic:
D = | Oi - P i | (14.16)
14.49
Th er e a r e t a bles of cr itical values for t h e Kom olgor ov-Sm irn ov Two-Sa m ple Test
wh ich a r e a fun ction of t h e n u m ber of in t er va ls, K (Sm ir n ov, 1948; Ma ss ey, 1951; Siegel,
1956; Ka n ji, 1993).
Illu st r a ti on
To illu st r a t e t h e t r ip len gt h com pa r is on , figu r es 14.15 t h r ou gh 14.18 sh ow t h e

r es u lt s for fou r differ en t im peda n ce m odels - an em pir ical im peda n ce fu n ction , a n ega t ive
exponen t ia l im peda n ce fu n ction , a t r u n cat ed n ega t ive exp onen t ia l im peda n ce fu n ction , a n d
a logn or m a l im peda n ce fu n ct ion . As seen , t h e fit of t h e em pir ica l im peda n ce fu n ct ion is n ot
pa r t icu la r ly good, bu t get s pr ogr essively bet t er wit h t h e t h r ee differ en t m a t h em a t ica l
fun ctions.
Th e best fit is clea r ly with t h e lognorm a l fu n ct ion . With t h ese pa r a m et er s (mea n

cent er = 6.0 m iles , st a n da r d d evia t ion = 4.7 m iles , coefficien t = 1, origin exponen t = 1, a n d
dest ina t ion expon en t = 1.06), th e Coinciden ce Rat io wa s 0.93.
Bu t , a ga in , t h is is ju st on e cr it er ia , a lbeit on e t h a t fit s m ost of t h e dis t r ibu t ion

m a t r ix. As w it h t h e n u m ber of in t r a -zon a l t r ips , m in im izin g t h e er r or for a t r ip len gt h
dis t r ibu t ion will n ot n ecessa r ily m in im ize t h e er r or for t h e ot h er t wo cr it er ia (in t r a -zon a l
t r ips a n d t h e t op link s). Bu t , its im por t a n t t h a t t h e t r ip len gt h com pa r ison be r ea son a bly
close.
Co m p a r i n g th e T ri p s o f t h e To p Li n k s
Th e t h ir d eva lu a t ion t est focus es on t h e t op link s. Th a t is, it eva lu a t es h ow well t h e

pr edict ed m odel ca pt u r es t h e m a jor t r ip lin k s, bot h in t r a -zon a l a n d in t er -zon a l. Sin ce
crim e t r ips a r e ver y sk ewed (i.e., a h a n dfu l of zones con t r ibu t e t o most crim e origin s a n d a
h a n dfu l of zon es a t t r a ct m a n y cr im es), ca pt u r in g t h e m ost im por t a n t lin k s is essen t ia l for a
good cr im e d is t r ibu t ion m od el. Th is is pa r t icu la r ly t r u e s in ce a m od el t h a t pr od u ces t h e
bes t fit for t h e overa ll tr ip len gt h dis t r ibu t ion m a y not ca pt u r e t h e t op link s ver y well.
Th er efor e, sim ply com pa r in g t h e t r ip len gt h dis t r ibu t ion m a y not a dequ a t ely
capt ur e the top links. That is, on a verage a par ticular model ma y produce a good fit
bet ween t h e pr edict ed a n d obs er ved dis t r ibu t ion s, bu t m a y d o t h is by m in im izin g er r or
a cr oss t h e en t ir e m a t r ix of t r ip pa ir s wit h ou t n ecessa r ily m in im izin g t h e er r or for t h e t op
link s.
Conse qu en t ly, it s im por t a n t t o als o comp a r e t h e fit of t h e m odel for t h e t op lin k s.

On e of t h e lin es in t h e dia logu e for t h e Comp a r e obser ved a n d p r edicted t r ip len gt h s is
Com p a r e t op lin k s. Th e u s er s h ou ld sp ecify t h e n u m ber of t op lin k s t o be com p a r ed ; t h e
defau lt is 100. Th e t op link s a r e t h e t r ip p a ir s t h a t h a ve t h e m ost n u m ber of a ctu a l t r ips ,
st a r t in g fr om t h e pa ir wit h t h e m ost t r ips a n d s ort in g in descen din g or der . Th e r out in e
ca lcu lat es a pseu do-Ch i squa r e t est on ju st t h ose link s. Since th e t op link s will a ll h a ve a
su fficien t n u m ber of t r ip s, it is possible t o ca lcu la t e a Ch i s qu a r e st a t is t ic. H owever , s in ce
14.50
Figure 14.15:
Comparing Observed and Predicted Crime Trip Lengths
Empirical Impedance Function
40%
Percentage of all crime trips
30%
20%
10%
0%
0 5 10 16 21 26 31
Trip length (miles)
Actual trips Modeled trips

Figure 14.16:

Negative Exponential Impedance Function
40%
30%
20%
10%
0%
0 5 10 16 21 26 31
Trip length (miles)

Figure 14.17:

Truncated Negative Exponential Impedance Function
40%
30%
20%
10%
0%
0 5 10 16 21 26 31
Trip length (miles)

Figure 14.18:

Lognormal Impedance Function
40%
30%
20%
10%
0%
0 5 10 16 21 26 31
Trip length (miles)

n ot a ll link s a r e being con sider ed in t h is t est , a sign ifica n ce t est of t h is st a t ist ic ca n n ot be
ca lcu lat ed since th e sa m pling err or is not kn own .
Us ing t h e obser ved (a ct u a l) lin ks a s t h e r efer en ce, th e t est ca lcu lat es:
K (P i O i )2
P seu do-ch i squa r e = G [-------------] (14.17)
i=1 Oi
wh er e P i is t h e pr edicte d n u m ber of t r ips for t r ip p a ir i, O i is t h e obser ved (a ct u a l) n u m ber

of t r ip s for t r ip pa ir i, a n d i is th e n u m ber of t r ip pa ir s t h a t a r e com p ar ed up to K
com pa r ison s, wh er e K is select ed by th e u ser .
Nu m ber of lin k s to test
The num ber of top links th at ar e to be compa red depends on h ow skewed is th e

dist r ibut ion . One good wa y to look a t t h is is t o plot t h e rank size d is t r ibu t ion of t h e
obser ved t r ips . Us in g t h e out pu t dbf file for t h e obser ved t r ip d ist r ibu t ion (see Ca lcula t e
obser ved or igin-dest in a t ion t r ips a bove), impor t t h e file int o a s pr ea ds h eet . Sort t h e file in
descend ing or der of t h e t r ip fr equ en cy an d crea t e a n ew var iable ca lled Ra n k ord er ,
wh ich is sim ply t h e descen din g or der of t h e t r ip fr equ en cies. Th en , p lot t h e fr equ en cy of
t r ips (FRE Q) on t h e Y axis a ga in st t h e r a n k ord er of t h e t r ip p a ir s on t h e X axis .
F igur e 14.19 below sh ows t h e r a n k size d ist r ibu t ion of t h e Ba lt im ore Coun t y cr im e
t r ips. Notice h ow t h e dist r ibut ion is very sk ewed for t h e t op crim e t r ip pa irs , but declines
su bst a n t ia lly a ft er t h a t . Th a t is , t h e t op t r ip lin k (wh ich wa s a n in t r a -zon a l t r ip pa ir - zon e
654 to itself) h a d 278 t r ips. The s econ d t op link (a lso a n int r a -zon a l pa ir - zon e 714 to
it self) h a d 226 t r ip s. Th e t h ir d h a d 223; t h e fou r t h h a d 205; a n d so for t h . As m en t ion ed
a bove, t h e t op 1000 t r ip lin ks a ccou n t for a bou t 47% of a ll t h e t r ip in t h e m a t r ix, but t h e
firs t 176 a ccoun t pa ir s a ccoun t for h a lf of t h a t . In oth er words , if t h e t op 150 t o 200 t r ip
pa ir s a r e exa m in ed, t h e h ighest volu m e lin k s will be in clud ed a n d m ost of t h e sk ewn ess in
t h e dist r ibu t ion will be a ccoun t ed for . Th e r em a in in g dist r ibu t ion , wh ich is n ot fitt ed, will
be less skewed.
Illu st r a ti on
An illustr at ion of how compa ring the top link s can modify a t rip distr ibut ion m odel
ca n be given. The s a m e m odel a s sh own in figur e 14.12 wa s r u n . The ps eu do-Ch i squa r e
t est for t h e firs t 176 pa irs wa s 5,832 (r ou n din g-off t o t h e n ea r est int eger). H owever, by
m odifyin g th e m ea n dist a n ce of t h e lognorm a l fu n ct ion a lower Chi squ a r e valu e wa s
obt a ined . Aft er severa l it er a t ion s, t h e lowest Chi squ a r e valu e wa s obta ined for a m ea n
dista nce of 5.2 miles (P2 = 5,448).
14.55
Figure 14.19:
Rank Size Of Observed Trip Distribution
300
250
Number of observed crime trips
200
150
100
50
0
0 100 200 300 400 500 600 700 800 900 1000
Rank order
Aga in , t h e t op lin k s r epr esen t s on ly on e cr it er ia ou t of t h e t h r ee m en t ion ed. A good
m odel sh ou ld bala n ce a ll t h r ee of t h ese.
Optimizing th e Three Evaluatio n Criteria
Th e ide a l solut ion would be t o ha ve a ll t h r ee eva lu a t ion crit er ia m in im ized. Th a t is,

wit h a n idea l m odel, t h er e sh ould be ver y litt le er r or bet ween t h e pr edicted m odel a n d t h e
obs er ved dis t r ibu t ion for t h e n u m ber of in t r a -zon a l t r ip s, t h e t r ip len gt h d is t r ibu t ion , a n d
th e top links.
In pr a ctice, it is u n lik ely t h a t a n y one m odel will m in im ize a ll t h r ee t ypes of er r ors .

Th u s, a ba la n ce (a com pr omis e) m u st be obt a in ed in ord er t o pr odu ce an opt im a l solu t ion.
Sin ce a ba lan ce ca n be obt a ined in differ en t wa ys, th er e a r e m u ltiple solu t ion s possible.
H in t: In Crim eSt a t , it is very eas y to ru n t h r ou gh differ en t m odels. The

pa r a m et er s a r e inpu t on t h e Set u p origin -dest ina t ion m odel pa ge. The
coefficien t s a r e calibr a t ed in t h e Ca libr a t e origin -dest in a t ion m odel r out in e
on t h e Or igin -Dest in a t ion Model pa ge. Th e coefficien t file wh ich is ou t pu t
is t h en in pu t in t o t h e Apply pr edict ed or igin -dest in a t ion m odel rou t in e on
t h e sa m e pa ge. Th e com pa r ison bet ween t h e obser ved a n d p r edicted va lu es is
fou n d in t h e Comp a r e obser ved a n d p r edicted origin -dest in a t ion t r ip
lengt h s r ou t ine. On ce set u p, iter a t ion s of t h e m odels can be ru n very
ea sily. A cha n ge is m a de on t h e set u p p a ge. Th e m odel is ca libr a t ed. It is
t h en a pp lied t o th e calibr a t ion d a t a set . Fin a lly, a com pa r ison is m a de.
Since the file nam es rema in const an t, an en tire itera tion t ak es less tha n a
m in u t e on a fa st compu t er (1.6 Gb or fa st er ).
To illu st r a t e t h e m u lt ip le cr it er ia , t a ble 14.2 sh ows t h e be st m odels for ea ch of th e

t h r ee t est s wit h va r ia t ion s on t h e m ea n dis t a n ce in t h e m odel sh own in figu r e 14.12. All
oth er pa r a m et er s wer e h eld con st a n t . Ma n y m odels w er e r u n t o pr odu ce th is t a ble
inclu din g tes t ing ot h er fu n ct ion s. Thes e a r e t h e t h r ee best .
As seen , d iffer en t m od els pr od u ce t h e lowes t er r or for ea ch of t h e cr it er ia . F or

obt a in in g t h e clos es t fit t o t h e n u m ber of in t r a -zon a l t r ip s, t h e m ea n d is t a n ce of t h e
lognorm a l fu n ct ion wa s 3.5 miles. For pr odu cing t h e best fit t o t h e t op 176 link s, t h e m ea n
dis t a n ce for t h e bes t m odel wa s 5.2 m odels. F or p r odu cing t h e bes t fit for t h e en t ir e t r ip
len gt h dis t r ibu t ion , t h e m ea n dis t a n ce of th e bes t m odel wa s 6.0 m iles. Th e qu est ion is
wh ich on e t o u se?
14.57
Ta ble 14.2
Mu l ti p le C ri t e ri a in S e l e c t i n g a D is t ri b u t io n F u n c t i o n
Logn or m a l fu n ct ion
St a n da r d deviat ion = 4.7 m iles
Coefficien t = 1
Or igin exponen t = 1.0
Dest ina t ion expon en t = 1.06
Num ber of Chi s qu are

Me a n In t r a -z o n a l fo r to p Co i n c i d e n c e
d i s ta n c e Tr ip s 1 76 Li n k s Ratio
Obser ved 8272 - -
6.0 5463 5814 0.93
5.2 6296 5777 0.87
3.5 8275 5986 0.74
O n e sol u t ion for op t i m i z i n g d e ci si on s
On e possible s olut ion is t o opt im ize in t h e followin g wa y:
1. If t h e t r ip dis t r ibu t ion m a t r ix is h igh ly sk ewed (wh ich will occu r wit h m ost
crim e da t a set s), t h en it s ess en t ia l t h a t t h e t op lin k s be r eplicat ed close ly.
Th is w ould t a k e pr iorit y over t h e secon d cr it er ia , wh ich is m in im izin g t h e
er r or for t h e t r ip len gt h d is t r ibu t ion , a n d t h e t h ir d cr it er ia , wh ich is
minimizing th e err or in predicting int ra -zona l tr ips.
2. Next fit th e model to minimize th e Chi squa re value for t he t op link s. In t he

exa m ple a bove, t h is wou ld be t h e t op 176 pa ir s. Typ ica lly, t h e m ea n dis t a n ce
h a s t h e biggest imp a ct for a lognorm a l or n or m a l fu n ct ion a n d t h is wou ld be
a djus t ed fir st . For a n egat ive expon en t ial fu n ct ion , th e expon en t h a s t h e
st r on gest im pa ct . F or a lin ea r fu n ct ion , t h e slop e h a s t h e st r on gest im pa ct
a n d for a t r u n ca t ed n ega t ive exp on en t ia l, bot h t h e pea k dis t a n ce, for t h e
n ea r dis t a n ce, a n d t h e exp on en t , for t h e fa r dis t a n ce, h a s t h e biggest im pa ct s
(see cha pt er 9). Aga in , t h e a im is t o pr odu ce th e Ch i squ a r e for t h e t op link s
with t h e lowest valu e.
4. Th en , wh ile t r ying t o ma in t a in a Ch i squ a r e va lu e a s close t o th is m in im a l

va lu e a s p ossible, a dju st t h e m odel t o min im ize t h e er r or in t h e t r ip len gt h
com pa r ison . In t h is ca se, t h e m odel wit h t h e h ighest Coinciden ce Rat io is
t h a t wh ich m in im izes t h e er r or. F or logn orm a l a n d n orm a l fun ction s, t h e
st a n da r d d evia t ion is t h e n ext pa r a m et er t o ad just . For a n ega t ive
expon en t ial fu n ct ion , th e coefficient sh ou ld be ad just ed n ext. F or a lin ea r
fu n ct ion , t h e in t er cep t wou ld be a dju st ed next a n d for a t r u nca t ed nega t ive
14.58
expon en t ial t h e slope would be a djus t ed n ext. Again , th e a im s h ou ld be to
obta in t h e h ighest Coinciden ce Rat io wit h out losin g t h e fit for t h e t op link s.
5. F in a lly, if it is possible, a dju st t h e exp on en t s of t h e or igin s a n d dest in a t ion s

a n d t h e ot h er pa r a m et er s (e.g., t h e coefficien t in t h e logn or m a l a n d n or m a l
dis t r ibu t ions ) to r edu ce th e er r or in t h e t ota l n u m ber of in t r a -zon a l t r ips .
Typica lly, however , th ese do not a lter t h e r esu lts very m u ch . They ca n be
th ought of as fine tun ing adjustm ents.
Notice tha t t his hiera rchy fits th e highest volume tr ip links first, then fits t he
overa ll t r ip len gt h dis t r ibu t ion, a n d fin a lly fits t h e n u m ber of in t r a -zon a l t r ips .
Illu st r a ti on
To illust r a t e, we firs t st a r t wit h t h e m odel t h a t pr odu ced t h e lowes t Ch i squ a r e.

Th a t m odel u se d a lognor m a l fun ction wit h a m ea n dis t a n ce of 5.2 m iles , a st a n da r d
devia t ion of 4.7 m iles, a coefficien t of 1, a n or igin exp on en t of 1.0 a n d a dest in a t ion
expon en t of 1.06. Va r yin g th e st a n da r d deviat ion of t h e lognorm a l fu n ct ion pr odu ced t h e
followin g r esu lt s (t a ble 14.3).
Ta ble 14.3
Mi n i m i z in g t h e S e c o n d Cr it e r ia i n S e l e c t in g a D i s t ri b u t io n F u n c t i o n
Mea n dist a n ce = 5.2 m iles
St a n da r d Deviat ion = 4.6 m iles
Coefficien t = 1
Num ber of Chi s qu are

Sta n da rd In t r a -z o n a l fo r to p Co i n c i d e n c e
d e v i a t io n Tr ip s 1 76 Li n k s Ratio
4.5 5809 5789 0.90
4.6 6057 5779 0.88
4.7 (ba seline) 6296 5777 0.87
4.8 6526 5780 0.86
4.9 6746 5788 0.84
As t h e s t a n da r d devia t ion wa s in cr ea s ed , t h e Coin cid en ce Ra t io d ecr ea s ed wh ile t h e

n u m ber of in t r a -zon a l t r ips in cre a se d. Of th es e five differ en t st a n da r d d evia t ions , 4.5
pr odu ced t h e h igh est Coin cid en ce Ra t io, bu t a ls o in cr ea sed t h e Ch i s qu a r e st a t is t ic for t h e
176 t op link s. Sin ce th a t crit er ia wa s s et fir st , we dont wa n t t o loosen it su bst a n t ia lly
du r ing t h e secon d a djus t m en t . Con sequ en t ly, a st a n da r d deviat ion of 4.6 was s elect ed
beca u se t h is in cre a se d t h e Coin ciden ce Ra t io sligh t ly wh ile n ot s u bs t a n t ia lly wor se n in g t h e
Ch i squ a r e t es t (an in cre a se of a bout 2).
14.59
Su bsequ en t t est s va r yin g t h e coefficien t of t h e logn or m a l fu n ct ion a n d t h e
exponen t s of th e origin a n d d es t in a t ion t er m s d id n ot a lt er t h es e va lu es . Cons equ en t ly, t h e
fina l model th a t wa s selected is list ed in t a ble 14.4.
Ta ble 14.14
Ba ltim ore Cou n ty Crim e Trips: 1993-1997

Optimal Mode l Selecte d
Mea n dist a n ce = 5.2 m iles
St a n da r d d evia t ion = 4.6
Coefficien t = 1
Th e m odel wa s r e-ru n wit h t h e n ew p a r a m et er s u sed. Th e t op 176 p r edicted t r ip

link s wer e ou t pu t a n d wer e com pa r ed t o t h e t op 179 obser ved tr ip lin ks (wh ich exceeded
176 beca u se of t ied valu es). The t op pr edict ed 176 link s a ccou n t ed for 7,241 tr ips, or 17.3%
of t h e t ot a l nu m ber of t r ips. The t op observed 179 link s a ccou n t ed for 9,900 tr ip, or 23.6%
of t h e t ot a l. Com p ar ed to t h e obs er ved dis t ribu t ion , t h e t op 176 p red ict ed lin k s a ccou n t ed
for a sma ller proport ion of th e tota l tr ips.
H owever, t h e fit wa s gener a lly bet t er . Figur e 14.20 sh ows t h e t op pr edict ed int er -
zon a l t r ip lin k s a n d com pa r es t h em t o t h e t op obs er ved lin k s wh ile figu r e 14.21 sh ows t h e
t op pr edicted in t r a -zon a l (local) tr ip lin k s a n d com pa r es t h em t o th e t op obser ved in t r a -
zon a l link s. Com pa r ing t h ese m a ps t o figu r e 14.12 a n d 14.13 (wh ich m a pped t h e t op 1000
lin k s, n ot t h e t op 176), th e fit is a bit bet t er for t h e m a jor link s, wh ich is wh a t we
opt imized. The fit is not p er fect ; it pr oba bly will n ever be. But , it is r ea sona bly close.
Of cou r s e, t h is is n ot t h e on ly wa y t o op t im ize a n d d iffer en t u s er s m igh t a p pr oa ch it

differ en t ly (e.g., min im izin g t h e in t r a -zon a l t r ips firs t , t h en t h e over a ll t r ip len gt h
dist r ibut ion , an d fin a lly t h e t op link s). It h a s t o be rea lized t h a t opt imizing in a differ en t
ord er will pr obably pr odu ce var ying r esu lt s; t h er e is n ot, u n for t u n a t ely, a sin gle opt im u m
solut ion t o t h ese t h r ee cr iter ia. Tha t is why it is import a n t t o explicitly define h ow an
opt im a l s olu t ion will be obt a in ed. In t h a t wa y, u ser s of t h e m odel ca n be cogn iza n t of
wh er e t h e m odel is m ost a ccu r a t e a n d wh er e it is pr oba bly less a ccu r a t e.
Im p le m e n ti ng th e Co m pa ri so n s in C r i m eS t a t
The mecha nics of condu cting the test s is fairly str aight forwa rd. The th ree tests a re
im plem en t ed in t h e Comp a r e Obser ved a n d P r edicted Tr ip Len gt h s r out in e on t h e la st
pa ge of t h e Trip d ist r ibut ion m odu le.
14.60
Figure 14.20:
Top Zone-to-Zone Trips from Optimized Model
All Crime Types
Baltimore County
Predicted top inter-zonal trips

25 or less
26 - 49
50 - 74
75 - 99
100 or more
Observed top inter-zonal trips
City of Baltimore 25 or less
26 - 49
50 - 74
75 - 99
100 or more
Baltimore County
City of Baltimore
N
0 10 20 Miles
W E
S
Figure 14.21:
Top Intra-zonalTrips from Optimized Model
#
All Crime Types
#
# #
Optimized intra-zonal predicted trips

#
#
#
Less than 50
#
Baltimore County # 50 - 99
#
#
#
100-149
#
#
#
# #
#
150-199
#
# #
# #
# #
#
# #
# #
# # #
# #
#
#
#
# #
# #
#
#
200 or more
# #
# #
#
#
# # #
# #
#
# #
# #
# # #
Top intra-zonal observed trips

#
#
#
# #
#
# #
#
Less than 50
#
# # # # #
#
#
# # # #
# # #
#
#
50 - 99
# # #
#
# #
#
#
# # #
# # #
#
#
# # # #
# #
# #
# # #
100-149
# # #
#
# # # #
# ## # #
# # #
# # # #
# # # # #
#
# # #
# # # #
#
# #
# # # # #
# # # # # # #
# # #
#
#
# # #
# # # # #
# #
# #
# # #
#
150-199
#
# # # # # # # #
#
#
##
#
# #
#
# # #
#
# #
# #
#
#
#
# # # #
## #
#
#
##
#
#
# #
#
# #
#
# # # #
#
##
# # # #
# # #
# # #
200 or more
#
#
#
#
#
# # ##
#
# # #
#
# #
# # #
#
##
#
##
# # #
# #
#
# # #
#
# #
# # #
# # #
# #
# #
Baltimore County
# #
# #
# #
# # #
#
City of Baltimore
#
#
# # # #
#
#
# # # # #
# #
# #
# #
### #
# #
##
# # # # #
# #
#
# #
#
#
#
#
#
# #
#
## # #
#
#
# #
# # #
# #
##
#
# # # # #
# #
#
#
# #
#
#
N
0 10 20 Miles
W E
S
O b s er v e d t r i p fi l e
Select t h e obs er ved t r ip dis t r ibu t ion file by click in g on t h e Br owse bu t t on a n d

find ing t h e file.
O b s er v e d n u m b er o f o r i g in -d e st i n a t i o n t r i p s
Specify th e var iable for t h e obser ved nu m ber of t r ips. Th e defa u lt n a m e is FRE Q.
O r i g _ID
Specify th e ID na m e for t h e or igin zon e. The defau lt n a m e is ORIGIN.
Note: th e ID s u sed for t h e origin zon es m u st be t h e sa m e a s in t h e dest in a t ion file

a n d t h e sa m e a s in t h e pr edict ed t r ip file if t h e t op link s a r e t o be com pa r ed.
O r i g _X
Sp ecify t h e n a m e for t h e X coord in a t e of t h e origin zon e. Th e defau lt n a m e is

ORIGINX.
O r i g _Y
Sp ecify t h e n a m e for t h e Y coord in a t e of t h e origin zon e. Th e defau lt n a m e is

ORIGINY.
D es t _ID
Specify th e ID n a m e for t h e dest ina t ion zon e. The defau lt n a m e is DES T.
n a m es a n d t h e sa m e a s in t h e pr edict ed t r ip file if t h e t op link s a r e t o be com pa r ed.
D es t _X
Sp ecify t h e n a m e for t h e X coord in a t e of t h e dest in a t ion zon e. Th e defau lt n a m e is

DE STX.
D es t _Y
Sp ecify t h e n a m e for t h e Y coord in a t e of t h e dest in a t ion zon e. Th e defau lt n a m e is

DE STY.
14.63
P r e d i c t e d t r i p fi l e
Select t h e pr edict ed t r ip dis t r ibu t ion file by click in g on t h e Br owse bu t t on a n d

find ing t h e file.
P r e d i c t e d n u m b er o f o r i g in -d e st i n a t i o n t r i p s
Sp ecify t h e va r ia ble for t h e obser ved n u m ber of t r ips . Th e defau lt n a m e is

PREDTRIPS.
O r i g _ID
Sp ecify t h e ID n a m e for t h e origin zone. Th e defa u lt n a m e is ORI GIN.
Note: th e ID s u sed for t h e origin zon es m u st be t h e sa m e a s in t h e dest in a t ion file

a n d t h e sa m e a s in t h e obser ved tr ip file if t h e t op link s a r e t o be com pa r ed.
O r i g _X
Sp ecify t h e n a m e for t h e X coord in a t e of t h e origin zon e. Th e defau lt n a m e is

ORIGINX.
O r i g _Y
Sp ecify t h e n a m e for t h e Y coord in a t e of t h e origin zon e. Th e defau lt n a m e is

ORIGINY.
D es t _ID
Specify th e ID n a m e for t h e dest ina t ion zon e. The defau lt n a m e is DES T.
n a m es a n d t h e sa m e a s in t h e obser ved tr ip file if t h e t op link s a r e t o be com pa r ed.
D es t _X
Sp ecify t h e n a m e for t h e X coord in a t e of t h e dest in a t ion zon e. Th e defau lt n a m e is

DE STX.
D es t _Y
Sp ecify t h e n a m e for t h e Y coord in a t e of t h e dest in a t ion zon e. Th e defau lt n a m e is

DE STY.
14.64
S el ec t b in s
Sp ecify how th e bin s (in t er va ls) will be defined. Th er e a r e t wo ch oices. On e is t o

select a fixed n u m ber of bin s. Th e oth er is t o select a con st a n t in t er va l.
Fi xed n u m ber
Th is set s a fixed n u m ber of bin s. An in t er va l is defin ed by t h e m a xim u m dis t a n ce

bet ween zone divid ed by t h e n u m ber of bin s. Th e defa u lt n u m ber of bin s is 25. Specify t h e
nu mber of bins.
C on s t a n t i n t e r v a l
Th is d efin es a n in t er va l of a sp ecific size. If select ed, t h e u n it s m u st a lso be ch osen .

Th e defa u lt is 0.25 m iles . Ot h er dis t a n ce un it s a r e n a u t ical m iles , feet , kilomet er s, a n d
m et er s. Specify th e int er val size.
C om p a r e t op l i n k s
Th e Com pa r e t op <va lu e> lin k s dia logu e im plem en t s a com pa r is on of t h e t op

link s. The u ser specifies t h e n u m ber of lin ks t o be com pa r ed. The defau lt is 100. The
rout ine calculat es a Chi squar e stat istic for t hese links.
Note: in or der t o m a ke t h e com pa r ison , th e or igin a n d des t ina t ion ID's m u st be th e

sa m e for both t h e obse r ved a n d p r edicte d t r ip files .
S a ve com p a r ison
Th e out pu t is s a ved a s a dbf file sp ecified by t h e u se r .
T a b l e ou t p u t
The table out put includes summ ar y inform at ion a nd:
1. Th e n u m ber of t r ips in t h e obser ved or igin-dest in a t ion file

2. Th e n u m ber of t r ips in t h e pr edicted origin -dest in a t ion file
3. Th e n u m ber of in t r a -zon a l t r ips in t h e obser ved or igin-dest in a t ion file
4. Th e n u m ber of in t r a -zon a l t r ips in t h e pr edicted origin -dest in a t ion file
5. Th e n u m ber of in t er -zon a l t r ips in t h e obser ved or igin-dest in a t ion file
6. Th e n u m ber of in t er -zon a l t r ips in t h e pr edicted origin -dest in a t ion file
7. Th e a ver a ge obser ved t r ip len gt h
8. Th e a ver a ge pr edicte d t r ip len gt h
9. Th e m edia n obser ved t r ip len gt h
10. Th e m edia n pr edicte d t r ip len gt h
11. Th e Coinciden ce Rat io (a n ind ica t or of con gru en ce var yin g fr om 0 to 1)
12. Th e D valu e for t h e Kom olgor ov-Sm irn ov two-sa m ple t est
14.65
13. Th e cr itical D value for t h e Kom olgor ov-Sm irn ov two-sa m ple t est
14. Th e p-va lu e a ss ocia t ed wit h t h e D va lu e of Komolgor ov-Sm ir n ov t wo-sa m ple
t est r elat ive t o t h e cr itical D value.
15. Th e pseu do-Chi s qu a r e t est for t h e t op link s
a n d for ea ch bin :
16. Th e bin n u m ber

17. Th e bin d is t a n ce
18. Th e obs er ved pr opor t ion
19. Th e pr edicted pr opor t ion
Fi l e ou t p u t
The saved file includes:
1. The bin num ber (BIN)

2. Th e bin dist a n ce (BIND IST)
3. The observed pr opor t ion (OBSERVPROP )
4. The pr edict ed pr opor t ion (PRE DPROP )
Gr a p h
Wh ile t h e ou t pu t pa ge is open , click in g on t h e gr a ph bu t t on will d is pla y a gr a ph of

t h e obs er ved a n d pr edict ed t r ip len gt h pr opor t ion s on t h e Y-axis by t h e t r ip len gt h dis t a n ce
on t h e X-a xis . Th is wou ld pr od u ce a s im ila r gr a p h t o t h a t s een in figu r e 14.15 t h r ou gh
14.18 above.
Use s of Trip Distribut ion Ana lysis
Th er e a r e a n u m ber of u se s for t h e t r ip d ist r ibu t ion a n a lysis . Fir st , for policing, a n

a n a lysis of t h e a ct u a l (obser ved) t r ip dist r ibut ion ca n be valu a ble. Secon d, th e pr edict ed
m odel h a s va lu e, a bove-a n d-beyon d t h e a n a lysis of t h e a ct u a l d is t r ibu t ion .
U t i li t y o f Ob s e r v e d Tr ip D i s tr ib u t i o n Ma p
Th is infor m a t ion by it self ca n be very us efu l for police. Two a pplica t ion s will be
discussed.
C r i m e p r e v en t i on effor t s
A ma jor a pplicat ion is using th e data shown in a t rip distr ibut ion m ap t o guide
en for cem en t effort s. F or exa m ple , in Ba lt im ore Coun t y, wit h t h e crim es occur r in g a t t h e
five sh oppin g m a lls, t h e origin loca t ions can be m ore ea sily s een . Th is h a s u t ilit y for police.
F ir st , t h e police int er ven e m ore e ffectively on t h e r out es lea din g fr om lik ely origin
loca t ion s. Th ey ca n pa t r ol t h os e r ou t es m or e h ea vily a n d, p er h a ps , in t er ven e m or e
14.66
fr equ en t ly. By u s in g t h e in for m a t ion fr om t h e t r ip dis t r ibu t ion a n a lys is , t h ey m a k e t h eir
en for cem en t effor t s sm a r t er . Secon d, t h ey ca n con du ct cr im e pr even t ion effor t s m or e
effectively. By kn owing t h e lik ely origin of offend er s, in t er ven t ion effort s in t h e origin
zon es m a y h ea d off som e of t h ese in cid en t s. P r ogr a m s su ch a s w eed -an d -seed a n d a ft er -
sch ool p r ogr a m s depen d on pr ovidin g a lt er n a t ive fa cilit ies for you t h , h opin g t o r edir ect
t h em t o m or e con st r u ct ive a ct ivities. Th ese facilities ca n be placed in loca t ion s wh er e
m a n y cr imes or igina t e.
Im p r o ve d J o u r n e y t o cr i m e a n a l y si s
A secon d a pplica t ion is in gues sin g th e likely or igin of a ser ial offen der . In cha pt er
9, t h eories of t r a vel beh a vior by a ser ia l offen der wa s d iscu ss ed. Th e r esu lt in g a n a lysis
(geogr a ph ic p r ofilin g, J ou r n ey t o cr im e a n a lysis ) u t ilized in for m a t ion on t h e dis t r ibu t ion of
incidents comm itted by th e offender. On t he oth er ha nd, the tr ip distr ibut ion pa tt ern seen
in figu r e 14.4 pr ovides a pr oba bilit y m a p of offen der loca t ion s a n d gives m or e in for m a t ion
t h a n wa s evid en t in t h e J ou r n ey t o crim e m odel. Th a t m odel a ssign ed a lik elih ood of th e
offen der livin g a t a loca t ion (t h e origin ) on t h e ba sis of t h e dist r ibu t ion of th e in ciden t s.
Th er e wa s n o ad dit ion a l infor m a t ion u sed a bout like ly or igin locat ion s. Th is t r ip
dis t r ibu t ion m a p, on t h e ot h er h a n d, p oin t s t o cer t a in zon es a s bein g t h e lik ely or igin for
offen s es com m it t ed a t t he m a jor d es t in a t ion loca t ion s . Th er e is m or e s t r u ct u r e in t h is
a n a lysis t h a n in t h e J ou r n ey t o cr im e logic.
On e can t h in k of t h is in t er m s of a qu a si-Ba yesia n a pp r oach t o gu essin g t h e lik ely

or igin of a n offen der . The geogra ph ic pr ofilin g/J ou r n ey to cr ime logic a ssu m es n o prior
probabilities. The only inform at ion t ha t is used is th e distr ibut ion of crimes comm itted by
a ser ia l offen der a n d a m odel of crim e t r a vel dist a n ce (es se n t ia lly, a n im peda n ce fu n ction ).
Th e t r ip d ist r ibu t ion m a p, on t h e oth er h a n d, p oin t s t o cer t a in locat ion s a s bein g t h e lik ely
or igin for in cid en t s . Ad m it t ed ly, t h is is ba s ed on a la r ge s a m ple of ca s es r a t h er t h a n on e
pa r t icula r ser ia l offen der . Bu t , t h e m a p p oin t s t o cer t a in pr ior pr obabilit ies for a n origin
loca t ion . If a n a n a lys t cou ld com bin e t h os e a p pr oa ch es - u s in g a p r ior p r oba bilit y m a p
a lon g wit h t h e dis t r ibu t ion of in cid en t s com m it t ed by a ser ia l offen der , a m or e r ea list ic a n d
a ccu r a t e gu es s a bout t h e offen der s r es ide n ce locat ion could be obt a in ed.
In ot h er wor ds, t h e em pir ica l d escr ip t ion of cr im e t r a vel p a t t er n s is u sefu l for

policing, a bove-an d-beyond a n y m odelin g t h a t is d evelope d.
U t i li t y o f P r e d i c t e d Tr ip D i s tr ib u t i o n Ma p
Th e m odel a ls o h a s a lot of u t ilit y for bot h policin g a n d cr im e a n a lysis . A n u m ber of

exa m ples will be given . F ir st , it ca n be u sed for fo re c as ti ng . By ca libr a t in g t h e m odel on
on e da t a set , it be a pplied t o a fu t u r e da t a set . As m en t ion ed in ch a pt er 12, m u ch of th e
popu la t ion a n d em ploym en t da t a t h a t for m t h e ba sis of a t r ip gen er a t ion m odel com es fr om
a Met r opolit a n P la n n in g Or ga n iza t ion (MP O). Most MP Os in t h e Un it ed St a t es a ls o m a k e
for eca st s of fu t u r e p op ula t ion a n d em p loym en t . Th os e for eca st s ca n be, in t u r n , con ver t ed
in t o for eca s t s of fu t u r e cr im e or igin s a n d cr im e d es t in a t ion s . Th u s , on t h e a s su m p t ion t h a t
t h e dist r ibut ion t r en ds will rem a in t h e sa m e over t ime, t h e t r ip dist r ibut ion m odel ca n be
14.67
a pplied t o t h e for eca st set of or igin s a n d dest in a t ion s. Th is cou ld a llow a n exa m in a t ion of
possible ch a n ges in t h e cr im e dis t r ibu t ion (a ssu m in g t h a t t h e fu t u r e for eca st s a r e cor r ect
a n d t h a t t h e t r ip d ist r ibu t ion coefficien t s r em a in con st a n t ).
Secon d, a m odel of crim e t r ip d ist r ibu t ion ca n be u se ful for m odelin g ch an ge s in

la n d u s e s. For exam ple, if a n ew sh oppin g ma ll is being plan n ed, on e ca n t a ke t h e
exist ing t r ip gener a t ion m odel a n d a djus t it t o fit t h e pla n n ed sit u a t ion (e.g., a ddin g 500
r et a il jobs t o t h e zon e in wh ich t h e m a ll is bein g develop ed). Th en , t h e t r ip gen er a t ion
model is re-ru n with t he new expected dat a, and t he tr ip distr ibut ion m odel is applied to
t h e pr edicted crim e origin s a n d cr im e dest in a t ion s. Th e r esu lt would be a m odel of like ly
cr im e t r ip s t o t h e n ew sh oppin g m a ll. Th is ca n be u sefu l t o th e m a ll developer s, t o fut u r e
bu sin esses , an d t o t h e police. If it t u r n s out t h a t t h e m odel for eca st s t h er e will be a
sizeable n u m ber of cr ime t r ips t o t h a t m a ll, t h en pr event ive a ct ion s can be developed
befor e t h e m a ll is bu ilt (e.g., impr ovin g secu r it y des ign in t h e m a ll; im pr ovin g t h e pa r k in g
lot a r r a n gem en t ).
Th ir d, a m odel of crim e t r ip d ist r ibu t ion ca n h elp in a n a lyzin g futu re

i nt e rv e n ti on s . F or exa m ple, in cr ea sin g p olice pa t r ols in a h igh cr im e a t t r a ct ion a r ea ca n
be exa m in ed a s t o possible effect iven es s befor e t a k in g t h e t r ouble t o reorga n ize
deploym en t . Or, a ddin g a n ew dr u g tr ea t m en t cen t er or a n ew you t h cen t er ca n be
m odeled a s t o its poss ible effect iven es s in cha n gin g t h e n a t u r e of crim e t r ips . Aga in , t h e
in pu t is a t t h e da t a level, wh ich a ffect s t h e t r ip gen er a t ion m odel. Bu t t h e t r ip dis t r ibu t ion
m odel is a pplied t o t h e n ew out pu t s fr om t h e t r ip gen er a t ion m odel. Th e a dva n t a ge of a
m odel is t h a t it explore s a set of in t er ven t ions wit h out h a vin g t o act u a lly h a vin g t o
im p lem en t t h em ; it s a t h in k in g t ool for p la n n in g ch a n ge.
F our t h , an d fin a lly, a cr im e t r ip d ist r ibu t ion m odel is h elpful in developin g c rim e
th e ory . As in dica t ed in cha pt er 11, th e t h eor y of cr ime t r a vel h a s been very elemen t a r y
u p t o now. Th e pr im a r y focu s of a n a lysis h a s been only on t h e dest in a t ion s a n d on t h e t r ip
lengt h s a s m ea su r ed by dist a n ce t r a veled. A t r ip dist r ibut ion m odel, on t h e ot h er h a n d,
a n a lyzes bot h t r ip dest ina t ion s a n d t r ip or igins , an d can inclu de a m or e sophist ica t ed
m ea su r e of imp eda n ce t h a n sim ple dist a n ce. Beca u se t h e a n a lysis is con du ct ed over a
la r ger a r ea (a ju r is dict ion or a m et r op olit a n a r ea ), t h e h ier a r ch y of cr im e t r ip s ca n be
a n a lyzed sim u lta n eou sly an d t h e int er a ct ion bet ween or igins a n d des t ina t ion s can be
exa m in ed . In sh or t , a cr im e t r ip dis t ribu t ion m od el is a qu a n t u m lea p in sop his t ica t ion
a n d com plexit y comp a r ed t o th e u su a l J our n ey t o cr im e t ypes of m odels. H opefully, it will
gen er a t e even m ore s oph ist icat ed t ypes of m odels.
At t h is poin t , we don t h a ve a n y exa m ples of t h e u se of a cr im e t r ip dis t r ibu t ion

m odel for policy in t er ven t ion or a n a lysis of la n d u se ch a n ges. Th e focu s h a s been on
pu t t ing t h e m odel t ogeth er a n d en su r ing t h a t t h e r ou t ines wor k p r oper ly. But , it is hoped
t h a t t h es e a pp licat ions will be pr es en t ed in fut u r e ver sion s of th e pr ogra m .
Th e n ext cha pt er con t in u es t h e t r a vel dem a n d m odel by exa m in in g how cr im e t r ip

lin k s a r e s plit in t o d iffer en t t r a vel m od es . Th a t is , t h e t r ip dis t ribu t ion m od el es t im a t es
14.68
t h e n u m ber of t r ips flowing fr om ea ch origin zon e t o each dest in a t ion zon e. Th e m ode sp lit
model th en breaks t hese trips int o distinct t ra vel modes.
14.69
En d n ot e s fo r Ch ap te r 14
1. Dis t a n ce ca n be u sed a s a r ou gh a ppr oxim a t ion for im peda n ce, bu t is r a r ely a good
pr edict or of a ct u a l tr a vel beha vior . For exam ple, in t h e m ode sp lit m ode t h a t will be
dis cus sed in cha pt er 15, t h e dist a n ce bet ween a locat ion a n d t h e n ea r est bu s or r a il
r ou t e ca n be us ed t o quickly select t r ip pa irs t h a t m igh t t r a vel by tr a n sit . However ,
t h e a ct u a l pr edict ion m u st be bas ed on a n et wor k calculat ion of t r a vel t ime or t r a vel
cost in t r a vers ing t h e syst em .
2. Most of t h e r esea r ch on fa ct or s a ffect ing u se of t r a n sit wer e con du ct ed in t h e 1960s

a n d 1970s. Thes e a ssu m pt ion s a r e m or e or less a ssu m ed by tr a vel dem a n d
modelers, rat her t ha n docum ented per se. See Schn ell, Sm it h , Dim sd a le, a n d
Th r a sh er , 1973; Roe m er a n d Sin h a , 1974; WAS H COG, 1974; Ca r n egie-Mellon
Un iver sit y, 1975; J oh n son, 1978; Levin e a n d Wa ch s, 1986b for some exam ples.
14.70

Trip Distribution: Theoretical Background

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Trip Distribution: Theoretical Background

Diunggah oleh

Hak Cipta:

Format Tersedia

Chapter 14

Tr ip dis t ribu t ion u su a lly occu r s t h r ou gh a n a lloca t ion m od el t h a t sp lit s t r ip s fr om

F or exam ple, wit h t h e Balt imore Coun t y dat a t h a t a r e being us ed t o illus t r a t e t h e

E a ch cell in t h e m a t r ix in dica t es t h e n u m ber of trips t h a t go fr om ea ch or igin zon e

In su ch a m odel, cons t a n cy is m a in t a in ed in t h a t t h e n u m ber of t r ips origin a t in g

wh er e t h e origin s, O i , a r e su m m ed over M origin zones w h ile t h e des t in a t ions , D j, a r e

Exam ple Crime Origin-De stin ation Matrix

Th e ba la n cing equ a t ion is im plem en t ed in a ser ies of st eps t h a t in clud e m odelin g

Th er e a r e t wo tr ip d ist r ibu t ion m a t r ices t h a t n eed t o be dist in gu ish ed. Th e firs t is

In ch a pt er 12, it wa s m en t ion ed t h a t t h er e is alwa ys un cer t a int y as t o t h e t r u e

Th u s, a m odel of t h e t r ip d ist r ibu t ion is p r odu ced t h a t a pp r oxim a t es t h e a ctu a l,

A m od el t h a t is u su a lly u s ed for t r ip dis t r ibu t ion is t h a t of t h e gravity fun ction , a n

wh er e g is a con st a n t or s calin g fa ctor wh ich en su r es t h a t t h e equ a t ion is ba la n ced in

Th e gra vit y model ha s been t h e bas is of m a n y app lica t ion s t o h u m a n societies a n d

wh er e I ij is th e int eraction between center s I an d j, P i an d P j ar e the respective populat ions,

St ewa r t (1950) a n d Zipf (1949) a pplied t h e con cept t o a var iety of ph en om en a

wh er e t h e t er m s a r e a s in equa t ion 14.3 but t h e expon en t of dist a n ce is on ly 1. Given a

In t h e St ewa r t /Zipf fr a m ewor k , t h e t wo Ps wer e both popu la t ion sizes . H owever , in

It sh ou ld be obviou s t h a t t h is in t er a ct ion equ a t ion ca n be a pplied t o t r ip s fr om on e

wh er e P i a r e t h e pr odu ction s for zone I, Aj a r e t h e a t t r a ction s zon e j, " is a pr odu ction

wh er e Aij is t h e n u m ber of pu r ch a ses in loca t ion j by r esid en t s of loca t ion I, S j is t h e

E qu a t ion 14.6 is a sin gle con st ra in t m odel in t h a t only t h e a t t r a ct iven ess of a

wh er e T ij is t h e in t er a ction bet ween t wo locat ions (or zones), P i is pr od uct ion s of t r ip s fr om

T i = " P i 8 G (AjJ /D ijD ) (14.8)

a n d gener a lizing t h is t o a ll zon es, we get:

On e of t h e pr oblems with t h e t r a dit ion a l gr a vit y for m u lat ion is in t h e m ea su r em en t

T ji = Aj$ e (-"D ij ) (14.10)

wh er e T ji is t h e a t t r a ction of loca t ion j for r es ide n t s of locat ion I , Aj is th e att ra ctiveness of

Note t h a t by m ovin g t h e dist a n ce te r m t o th e n u m er a t or, st r ictly spea k in g it n o

Th er e a r e ot h er dis t a n ce fu n ct ion s, a s well. Ch a pt er 9 exp lor ed som e of t h ese. F or

On e of t h e biggest a dva n ces in t h is t ype of model h a s been t o in crea se t h e flexibility

Th is r ea liza t ion h a s led t o t h e concep t of t ra vel im p ed a n ce which, in t ur n, has led to

Default Home-Based Work Trip Impedance

If t r a vel t im e is a crit ical com pon en t of t r a vel, wh y t h en don t offend er s com m it

An even bet t er con cep t of im p ed an ce is th a t of trav el cost (somet imes ca lled

Bu t cos t in volves ot h er d im en s ion s t h a t n eed t o be con s id er ed . Th er e a r e r ea l

If t h e con cept is exp a n ded t o t h a t of a n offen der , t h er e a r e ot h er per ceived cost s t h a t

Th u s, in sh ort , a m ore r ea list ic m odel of t h e in cent ive or d isin cent ive t o ma k e a t r ip

Th e fin a l con cep t t h a t is in t r odu ced in defin in g im peda n ce is t h a t of travel utility.

But, for completeness sake, we need to un dersta nd t ha t t he likelihood or

wh er e t h e n u m ber of t r ip s fr om zon e i t o a ll ot h er zon es, j, is a fu n ct ion of t h e pr odu ct ion s

T ij = " P i 8 $ AjJ I ij (14.12)

wh er e P i is t h e pr odu ction cap a city of zon e I, Aj is t h e a t t r a ction of zone j, I ij, is a

Th er e a r e a lt er n a t ive a lloca t ion s pr ocedu r es t o t h e gr a vit y m odel. On e well kn own

wh er e Aji is t h e a t t r a ction of loca t ion j by r es ide n t s of locat ion I , S j is th e att ra ctiveness of

In pr a ctice, in sp it e of it s m ore in t u it ive t h eore t ical m odel, t h e in t er ven in g

An ot h er a lt er n a t ive m et h od wa s con d u ct ed by P or oja n (2000) in a p plyin g t h e

Con st r a in dest in a t ion s

Con st r a in bot h or igin s a n d dest in a t ion s

3. Th e r ou t ine t h en su m s t h e r ows a n d colum n s in t h e m a t r ix. Then ,

4. Th e r ou t in e r epea t s st eps 2 a n d 3 u n t il t h e ch a n ges fr om on e it er a t ion t o t h e

5. Th e la st es t im a t e of " or $ (or bot h for a double-con st r a ined m odel) is t a ken

6. On ce t h e pa r a m et er s h a ve been est im a t ed, t h e m odel ca n be a pplied t o t h e

7. A com p a r is on ca n be m a d e bet ween t h e obs er ved dis t r ibu t ion a n d t h e

Cr i m eS t a t III Tri p D is tri bu t io n Ro u ti n e s

Next , we exa m ine t h e a ct u a l tools t h a t a r e a vailable in t h e Crim eS tat t r ip

3. S e tu p ori gi n-d e st in a ti on m o de l. This screen a llows t h e u ser t o defin e t h e

4. Ca li bra te o ri gi n-d e st in a ti on m o de l. This rout ine calibrat es the

5. Ap ply pre d ic te d ori gi n-d e st in a ti on m o de l. This rout ine applies th e

E a ch of t h ese r ou t ines a r e described in det a il below. Figur e 14.3 shows a screen

An em p ir ica l d es cr ip tion of t h e a ct u a l t r ip dis t ribu t ion m a t r ix ca n be m a de if t h er e

The steps in r un ning the model ar e as follows:

1. Ca lc u l a te o b s e r v e d o r ig i n -d e s t i n a t io n t r ip s . Ch eck if a n em pir ica l

Lin ea r : f(d ij) = 0 + $S ij = $d ij for S ij $ 0, d i j# d p (14.19)