Article information:
To cite this document:
Barbara Morris Robert Johnston, (1987),"Dealing with Inherent Variability: The Difference Between
Manufacturing and Service?", International Journal of Operations & Production Management, Vol. 7 Iss
4 pp. 13 - 22
Permanent link to this document:
http://dx.doi.org/10.1108/eb054796
Downloaded on: 11 April 2015, At: 04:39 (PT)
References: this document contains references to 0 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 938 times since 2006*
Access to this document was granted through an Emerald subscription provided by 357736 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald
for Authors service information about how to choose which publication to write for and submission
guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
13
Books, journals and teaching curricula are currently using the phrase 'The Service Industries'
and separating them from 'The Production Industries', implying that there is a significant
difference in the management of the transformation process in the two types of industry[l].
This paper attempts to answer the question of whether there really is a difference
between service and manufacturing industries, and if so, what that difference is.
The case against there being a significant difference is strong. Most
production/operations management texts recognise that the concepts, tools and
techniques are as applicable to service as to manufacturing. Hill[2] for example states
"Throughout this book, therefore, the terms production management and service
operations management will be treated as being synonymous". Lockyer[l] specifically
argues that there is no difference, taking several indicators that authors have used to
distinguish service from manufacturing, such as the intangibility of products,
perishability, customer involvement in the process, and the back room/front office
split, and ably demonstrates that they are all applicable to manufacturing industries.
The case for there being a difference is weak, and seem to rest upon simply held
beliefs that there is a difference. Common sense suggests that this is so. Most people
would agree that Massey Ferguson is primarily a manufacturing organisation and the
leisure complex at Alton Towers is a service organisation. Because they are identified
as different types of organisation, common sense says that they are substantively
different and ipso facto will be managed differently.
Lockyer avoids this issue, and concludes that the only real concern is whether an
organisation is "for-profit" or "not-for-profit". Whilst we would agree that profitmaking and non-profit-making organisations may deserve different treatment, this
argument does not in itself deal with the issue of what, if any, is the difference between
service and manufacturing operations, and whether any difference between the two
is such that they are managed in different ways and therefore need different treatment
in the classroom.
We believe that at least part of the difficulty stems from the word "service". It has
many connotations[3]; in some texts it is used to embrace all non-manufacturing
operations, whilst in others it is used to refer to some kind of "pure" service, which
by implication is some kind of personal treatment. A number of the arguments about
whether or not services are different appear to be confused by differing under-
14 IJOPM 7,4
standings of what is meant by service. The key to this issue, however, does not, it seems
to us, lie in a debate about semantics, nor in trying to tighten up current definitions
of service, nor in testing these inadequate definitions against dubious criteria. Rather
it seems to lie, as Lockyer suggests in his opening sentence[1], in looking at the
transformation process itself.
Wild[4] has distinguished manufacturing, supply, transport and service operations,
but work by Morris[5] shows that whilst these categories usefully distinguish many
operations, several "grey areas" remain, where there can be overlap. We believe that
a distinction in terms of what is being transformed, rather than the type of
transformation, is rather more useful.
We propose that there are three distinct types of operation (Figure 1), characterised
by the inputs which are processed instead of the outputs they produce: customer
processing operations (CPO); information processing operations (IPO); and material
processing operations (MPO).
The CPO[6] acts upon the customer who enters the system, receives some kind of
treatment in the process or participates in the process, and who is the major output
from the system. The MPO acts upon input materials and transforms them into goods.
The IPO deals with information and converts it into a form which is usable either
by the system itself, or some other system. Typical outputs might be orders for a CPO
or MPO, drawings, or analyses to be used for control purposes.
Each operation, whether it processes customers, information or materials, has
facilitating goods, equipment and manpower and each has to deal with issues like
capacity planning, operations planning and control, management of inventories or
queues, maintenance of equipment, control of quality, etc.
Manufacturing organisations are predominantly MPOs, usually with several IPOs
which deal with customer orders, product and process specifications, material
requisitions, etc., and may include CPOs, particularly in a make to order system. Service
organisations involve customers, and may either process them in a CPO, or process
information about them or their order(s) in an IPO, and occasionally may process
Inherent Variability
15
material in an MPO. Despite this predominance of one type of operation rather than
another, most organisations will have examples of all three types.
The arguments about whether there are significant differences between
manufacturing and service seem to be based on discussions of whether there is anything
significantly different per se about the two types of industry. We would argue that
in essence there is no distinction between the two because the nature of the business
does not discriminate. The issue for operations management is not whether there is
anything different about the type of industry or business, but whether there is anything
singificantly different about processing materials, information or customers, and in
this paper we explore what we regard as the two extremes of material and customer
processing.
Clearly, there is much that is common across all operating systems this is the
fundamental premise underlying operations management and at the macro level
they all have similar tasks and problems. However, we argue that at the micro level
of an operation the nature of the problems varies, and different tools and techniques
are applicable. There is nothing new in this. It is well demonstrated in the traditional
distinction between job, batch and flow production. All are manufacturing systems
and have some common problems but they are distinguished by the variety and volume
of products made, and this in turn means that some tools and techniques are applicable
in only one of the types.
However, the impact of the distinction is felt largely at the planning stage. Once
a product is actually being processed by any of these types of system, then there is
a fairly high degree of commonality of management task. There is normally high
certainty about what is needed (the specification) and reasonable knowledge about
what will be the outcome from each stage in the process. The fundamental task,
therefore, is to ensure that the final outcome matches the specification. This aim is
achieved by controlling variability. Efforts are made to reduce material and process
variability as much as possible and then to control them within defined limits so as
to produce an output which meets the specification. This task of reducing and
controlling variability within the process remains the same regardless of the type of
operating system.
We consider that this is not the case in many customer processing operations. Here
the variability of both the transformed and the transforming resources are relatively
higher because human beings are the transformed input, and their performance and
output cannot be fully specified. This variability and uncertainty means that there
may be a need to increase, or at least maintain, variability in process capability in
order to cope with the variable input. In CPOs where this is the case, some of the
traditional manufacturing concepts are inapplicable the most obvious example being
the idea of quality improvement through continued reduction in variability.
A Comparison of Material and Customer Processing Operations
If we contrast MPOs and CPOs we can identify two production dimensions the
product or service itself as a desired output, and the processes used to produce it.
Within each of these dimensions are two aspects, those which are tangible and those
which are intangible.
16
IJOPM 7,4
Tangible aspects are those which can be specified objectively: they may be aspects
of the product or service or of the process as shown in Figure 2.
Intangible elements are those aspects which cannot be specified, except in very general
terms, and which are normally concerned with the customer's reaction to some element
of the product or process. For example in an MPO an intangible product element
would be the value the customer attaches to owning the final product, or his/her view
of whether or not the quality is acceptable. In a CPO intangible product elements
would be, for example, the customer's satisfaction with a haircut or the enjoyment
of a meal. These are shown in Figure 3.
As far as the tangible elements are concerned, there is no real difference between
an MPO and a CPO. In an MPO the customer either specifies what is required of
the material output from the system, or buys a product whose specification is
determined by the system which is believed to meet his/her requirements. Similarly
in a CPO any new customer arrives at the point of entry to the system with a set
of tangible requirements for the outcome of the system. These can be tested against
the tangible elements being offered by the system, or can be specified and, as with
the MPO, the customer can accept or reject what is offered. For example, a customer
requiring a haircut or a meal can check the delivery time and decide whether it is
acceptable just as can the purchaser of manufactured goods.
However, each customer will also have a set of intangible requirements which may
exist at a range of levels of consciousness. These form part of his/her perception of
an acceptable product or service a "pleasant experience", and "efficient service",
a product which is "good value", etc. The problem with these is twofold:
Inherent Variability
17
(1) The customer can rarely articulate them, and often has no opportunity to do
so. The provider therefore bases the product or service on his/her view of what
the customer's reaction will be.
(2) The perception of an acceptable product or service with respect to the intangible
elements varies from one customer to another.
The end result is that whilst tangible elements can be specified, and hence there is
a shared expectation about what will be provided, with intangible elements this is not
possible in advance of the receipt of the product or service. With repeat business there
is the opportunity for customer and provider to learn what each expects of the other,
but for the first-time purchaser of a product or service there is no way of sharing
completely the expectations of what is required and what will be provided. Advertising
and preliminary neogitations may reduce some of the uncertainty, but they can never
completely overcome it.
This issue of intangible requirements applies equally to both MPOs and CPOs
with respect to the product or service output which is provided, but there are
similar problems with the processing technology which do not apply equally.
With material processing, the variability of both the material and the process
capability is normally known and is controllable within acceptable limits. Hence
it is possible to predict within limits the outcomes of differing combinations
of materials and processes. Failures will occur more or less frequently because
either or both the material and the process do not conform to the pattern of variability
expected, but these are tangible variances and the failure itself is normally tangible.
Occasionally failure will occur because the operator does not behave in the expected
manner, but again this will have a tangible effect on the product. The customer may,
on receipt of the item, be dissatisfied with some intangible element, but this failure
occurs outside the system. The only failures occurring within the operation are related
to tangible elements. The material itself has no emotive response to what is done to
it, and hence no failure can be attributed to the material's "feelings" about whether
it is acceptable or not to wait for processing, or its objections to the mood of the
operator.
This is clearly not the case with customer processing. Here there is similar variation
in the tangible aspects of customer and process variability, but there is the additional
problem of the emotive response of both the customer and the process provider,
and the interaction between them. Failures can occur with the tangible elements
for the same reasons as failures in material processing operations, but in addition
there is considerable scope for failures relating to a mismatch between intangible
elements of the processing itself and what the customer expects or requires. For
example, a nervous or anxious patient is unlikely to be satisfied with a treatment which
deals with the medical problem but which does not also provide sympathy and
reassurance.
These failures with respect to intangible elements of the process arise because of:
(1) differing perceptions of what is needed, based on varying expectations; or
(2) adverse emotive reaction to the mode of processing.
18
IJOPM 7,4
The net result of all this is that whilst material and customer processing are very similar
with respect to tangible dimensions, differences exist with respect to intangible
dimensions.
In both types of operation there are intangible aspects of the output from the
operation to which the customer responds. The difference between the two types is
that in an MPO the customer receives the output after processing, and hence the
reaction is not felt directly during the process itself, whilst in a CPO the customer
receives the output while it is being produced, and not at the end of the process. Hence
the reaction is felt immediately and may have an impact on later stages in the
production process. Customer reaction to the output from an MPO can certainly have
an impact, but there is usually some lag in the system, and the MPO is usually buffered
from direct customer reaction by other systems such as sales. Hence an MPO does
not normally have to respond in real time during the operation itself, and reacts through
design and planning changes.
With respect to processing, the two types of operation are totally different. In MPOs
there is no intangible element because the material being processed is inanimate. In
CPOs there is an immediate reaction to intangible elements of processing and the
operation has to respond to this. The nature of these elements, and the response
required of the operation, is summarised in Figures 4 and 5.
Inherent Variability 19
Thesefiguresdemonstrate that whilst there are very similar characteristics with respect
to tangible elements, there are dissimilarities with respect to intangible elements, which
are essentially related to uncertainty, freedom of choice, and the timing of the reaction
needed. In MPOs there is rather more freedom of choice, less uncertainty, and time
lags and buffers in the wider system which obviate the need to respond during the
processing of a given piece of material.
Implications of Differences
Given this, two extreme choices are available to CPOs to become more like MPOs
by removing the uncertainty and the need for immediate responses to demands made
during processing, or to cultivate a flexible response which implies a heavy dependence
on diagnostic skills within the operation itself and the deliberate cultivation of highly
variable processing capability.
Current POM concepts, such as inventory management, capacity planning and
scheduling, and tools and techniques such as line balancing and queueing theory, deal
essentially with the tangible elements. It is not surprising, therefore, that there is
20 IJOPM
7,4
Inherent Variability
21
These three basic strategies require quite different customer management. The
significant differences are set out in Table I.
Table I. Comparison of Strategies for Dealing with Variability
Customisation
of Process
Standardisation
of Process
Surrogate Tangibles
Diagnosis of customer
needs within the process
System-specified provision,
based on perceived market
demand
Routine process
Routine process
These represent "pure" strategies, and few systems will employ any one of these to
the total exclusion of others, but many examples exist of systems in which one of them
predominates.
The approach advocated by Levitt[8] and epitomised by the so-called fast food
operations is clearly representative of operations where Strategy 2 predominates. In
personal services, such as career counselling, Strategy 1 normally predominates.
Strategy 3 is often the hallmark of professional services such as auditing or
management consultancy.
Generally, we would argue that the three strategies represent points on a continuum
of variable response to the management of intangibles, and that it is the position on
this continuum which determines similarity or difference between varying types of
operating system.
Strategy 3 represents a mode of operation which treats customer processing as if
it were raw material processing. A service operation which adopts this strategy is, in
our view, very similar to a manufacturing operation which makes products to order.
88
IJOPM 7,4
19. Henrique Luiz Corra, Lisa M. Ellram, Annibal Jos Scavarda, Martha C. Cooper. 2007. An
operations management view of the services and goods offering mix. International Journal of
Operations & Production Management 27:5, 444-463. [Abstract] [Full Text] [PDF]
20. Sanjib Chowdhury, Grant Miles. 2006. Customer-induced uncertainty in predicting organizational
design: Empirical evidence challenging the service versus manufacturing dichotomy. Journal of
Business Research 59, 121-129. [CrossRef]
21. Robert Johnston. 2005. Service operations management: return to roots. International Journal of
Operations & Production Management 25:12, 1278-1297. [Abstract] [Full Text] [PDF]
22. AFONSO FLEURY, MARIA TERESZA FLEURY. 2003. THE EVOLUTION
OF
STRATEGIES
AND
ORGANIZATIONAL
COMPETENCIES
IN
THE
TELECOMMUNICATIONS INDUSTRY. International Journal of Information Technology &
Decision Making 02, 577-596. [CrossRef]
23. Kee-hung Lai, T.C.E. Cheng. 2003. Initiatives and outcomes of quality management implementation
across industries. Omega 31, 141-154. [CrossRef]
24. Gavin Dick, Kevin Gallimore, Jane C. Brown. 2002. Does ISO 9000 accreditation make a profound
difference to the way service quality is perceived and measured?. Managing Service Quality: An
International Journal 12:1, 30-42. [Abstract] [Full Text] [PDF]
25. Tonya Boone, Ram Ganeshan. 2001. The effect of information technology on learning in
professional service organizations. Journal of Operations Management 19, 485-495. [CrossRef]
26. Gavin Dick, Kevin Gallimore, Jane C. Brown. 2001. ISO 9000 and quality emphasis An empirical
study of frontroom versus backroom dominant service industries. International Journal of Service
Industry Management 12:2, 114-136. [Abstract] [Full Text] [PDF]
27. Stuart Chambers, Robert Johnston. 2000. Experience curves in services: macro and micro level
approaches. International Journal of Operations & Production Management 20:7, 842-859. [Abstract]
[Full Text] [PDF]
28. Charlene Pleger Bebko. 2000. Service intangibility and its impact on consumer expectations of service
quality. Journal of Services Marketing 14:1, 9-26. [Abstract] [Full Text] [PDF]
29. Rhian Silvestro. 1999. Positioning services along the volumevariety diagonal. International Journal
of Operations & Production Management 19:4, 399-421. [Abstract] [Full Text] [PDF]
30. Robert Johnston. 1999. Service operations management: return to roots. International Journal of
Operations & Production Management 19:2, 104-124. [Abstract] [Full Text] [PDF]
31. Bart Van Looy, Paul Gemmel, Steven Desmet, Roland Van Dierdonck, Steven Serneels. 1998.
Dealing with productivity and quality indicators in a service environment: some field experiences.
International Journal of Service Industry Management 9:4, 359-376. [Abstract] [Full Text] [PDF]
32. Csar Camisn, Marisa Flor, Sonia Cruz, Ins Kster. 1996. Quality practices and perceptions of
Valencian hospitality enterprises: an empirical analysis. International Journal of Quality & Reliability
Management 13:7, 79-92. [Abstract] [Full Text] [PDF]
33. Christine A. Witt, Alan P. Muhlemann. 1995. Delivering a quality service in air travel: The problems.
The Tourist Review 50:1, 32-37. [Abstract] [PDF]
34. CA Witt, AP Muhlemann. 1994. The implementation of total quality management in tourism:
some guidelines. Tourism Management 15, 416-424. [CrossRef]
35. Jon Sundbo. 1994. Modulization of service production and a thesis of convergence between service
and manufacturing organizations. Scandinavian Journal of Management 10, 245-266. [CrossRef]
36. John A. Dotchin, John S. Oakland. 1994. Total Quality Management in Services. International
Journal of Quality & Reliability Management 11:3, 27-42. [Abstract] [Full Text] [PDF]
37. John A. Dotchin, John S. Oakland. 1994. Total Quality Management in Services. International
Journal of Quality & Reliability Management 11:3, 9-26. [Abstract] [Full Text] [PDF]
38. Robert Johnston, Richard Bryan. 1993. Products and Services - A Question of Visibility. The Service
Industries Journal 13, 125-136. [CrossRef]
39. Sue Perrott Siferd, W.C. Benton, Larry P. Ritzman. 1992. Strategies for service systems. European
Journal of Operational Research 56, 291-303. [CrossRef]
40. John A. Dotchin, John S. Oakland. 1992. Theories and concepts in total quality management. Total
Quality Management 3, 133-146. [CrossRef]
41. BinShan Lin. 1991. Building Quality Control Systems for Hospital Foodservice Operations.
International Journal of Quality & Reliability Management 8:2. . [Abstract] [PDF]
42. Bob Brotherton, Mike Coyle. 1990. Managing Instability in the Hospitality Operations
Environment. Part One: Variability. International Journal of Contemporary Hospitality Management
2:3. . [Abstract] [PDF]
43. Barbara Morris. 1989. Customer Satisfaction and Resource Productivity in Health Care. International
Journal of Health Care Quality Assurance 2:3. . [Abstract] [PDF]
44. John HaywoodFarmer. 1988. A Conceptual Model of Service Quality. International Journal of
Operations & Production Management 8:6, 19-29. [Abstract] [PDF]
45. Barbara Morris. 1988. Accommodating Multiple Objectives in the Design of Customer Treatment
Operations. International Journal of Operations & Production Management 8:3, 86-94. [Abstract]
[PDF]