Available online at www.sciencedirect.

com

Procedia - Social and Behavioral Sciences 40 (2012) 41 51

An Inventory Model for Deteriorating Items in a Supply Chain with System Dynamics Analysis
Ching-Fang Leea , Chien-Ping Chungb
ab
Department of Business Innovation and Development, Ming Dao University, Changhua, Taiwan

Abstract Rau et al. (2004) developed an integrated inventory model of supply chain for deteriorating items among a supplier, a producer and a buyer. Through mathematical model development, the optimal order lot-size and delivery number n was solved under the minimal joint cost of supply chain. Rau et al. emphasized on traditional mathematical and statistical calculation to acquire the optimal order timing and best inventory policy, but not had further discussion on dynamics complex relation due to time evolution. In order to reflect the operation process truly, there is a forceful need to consider inventory model of deteriorating items in a systematically way. Therefore, in this paper, we apply system dynamics thinking to propose a new order system and conduct a systematical simulation. The results of validation and model testing of constructed model showed that system dynamics simulation methodology is appropriate. Finally, through the system dynamics simulation, the optimal minimal joint cost of supply chain was acquired.

Elsevier Ltd. Selection and/or peer-review under responsibility of the 2012 2012Published Publishedby by Elsevier Ltd. Selection and/or peer-review under responsibility ofAsia the Pacific Asia Business Innovation and Technology Management Society Pacific Business Innovation and Technology Management Society (APBITM).
Key words: System dynamics; Supply chain; Deteriorating inventory

Dr. Ching-Fang Lee Department of Business Innovation and Development, Ming Dao University 369 Wen-Hua R.d., Peetow, Changhua, 52345, Taiwan Tel: +886-4-8876660 ext.7524 Fax: +886-4-8879036 Email: chinigfanglee@mdu.edu.tw

1877-0428 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Asia Pacific Business Innovation and Technology Management Society doi:10.1016/j.sbspro.2012.03.159

42

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

1.

Introduction

In order to fulfill the needs of market and customers, to manage inventory properly has become an important operation activity and a source of profits (Chorpa and Menidl, 2001; Minner, 2003; Ranjan and Susmita, 2011). Proper inventory management has great influence on cost down and production stability for companies (Krajewski et al., 2010). In real world, the inventory of many products may face limitations as deterioration, deadlines of consumption (DOC), and vaporizing. That can be called as deteriorating inventory. When developing inventory model which have the traits of deterioration, it is very important to understand the traits and categories of deterioration (Wee, 1999; Alamri and Balkhi, 2007; Hsu et al., 2010; Yi et al., 2010). Deterioration means the traits of wear away, lower utility and loss due to physical depletion phenomenon of objects or usage. Following by Ghare and Schrader (1963), deterioration can be divided into three categories: (1) direct spoilage: such as vegetables, fruits, seafood, and foods easily decayed. (2) physical depletion: such as gasoline and vaporizing liquids. (3) deterioration: such as a pharmaceutical expired, and electronic devices utility loss over time. From literature review, we realized that most scholars used mathematics calculation to find the approximate optimal solution when conducted inventory related researches. They did not have further discussion about dynamics complex related to time evolution (Rau et al., 2004; Ping et al., 2007; Kung et al., 2010; Yong et al., 2010). In this paper, based on the model of Rau et al. (2004), we proposed a new order system of inventory model. Then, we constructed a system dynamics model and conduct model testing and simulation. We hope that through constructing an analytical system dynamics model to solve the dynamics complex issues of inventory model of supply chain for deteriorating items. This study is written in six sections. Section 1 is the introduction. Section 2 describes the assumptions and notations in this study. Section 3 describes the study of Rau et al. (2004). Section 4 presents the system dynamics model. Section 5 provides model analysis and numerical example for applications. Finally, a conclusion is given in Section 6. 2. Assumptions and notations

2.1 Basic Assumptions The inventory model was developed based on the following assumptions: 1. Order cycle is known. 2. A single supplier, a single producer and a single buyer are considered. 3. Single item is considered. 4. Raw materials to finish goods factor is 1:1. 5. Demand rate and production rate are deterministic and constant. 6. Production rate is greater than the demand rate. 7. Deterioration rate is deterministic and constant. 8. Without shortages. 9. Multiple lot-size deliveries per order are considered instead of a single delivery per order. 10. Lot-size is the same for each delivery. 2.2 Notations The notations used in this study are showed as follows:
Symbol T n t n* For a buyer D Demand rate of finished goods for a buyer (constant) Deterioration rate of finished goods for a buyer (constant) Denotation Order cycle Number of deliveries per order cycle T Delivery cycle Optimal delivery number

tB
WB

Inventory consumption period for a buyer Inventory quantity on hold of finished goods from time 0 to time t B

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51
qB AB FB HB PB TCB For a producer P Production rate of finished goods for a producer (constant) Deterioration rate of finished goods for a producer (constant) tP qP WP SP FP HP PP TCP For a producers warehouse Deterioration rate of producers raw materials (constant) qPW WPW FPW HPW PPW TCPW For a supplier Deterioration rate of suppliers raw materials (constant) qS QS WS AS FS HS PS TCS Stock items TC S_inventory P_warehouse P_inventory V_shipping Others(converter) S_order_signal S_order_ point S_shipping PW_order_signal PW_order_ point Order signal for a supplier Reorder point for a supplier Total quantity of raw materials receiving for a producer Order signal of raw materials for a producer Reorder point of raw materials for a producer Joint total cost with TCB, TCP, TCPW and TCS Total inventory for Supplier Total raw material inventory for a producer Total finished goods inventory for a producer Total quantity of raw materials receiving from a suppliers vendor Quantity of raw materials per delivery from outside vendor to supplier Total quantity of raw materials receiving (planned) from a suppliers vendor during period T Inventory quantity on hold of raw materials for supplier from time 0 to t Ordering cost of raw materials per order for supplier (constant) Delivery cost of raw materials per delivery for supplier (constant) Holding cost of raw materials per unit per unit time for supplier (constant) Cost of a deteriorated unit of suppliers raw materials (constant) Total cost for supplier Quantity of raw materials per delivery from a supplier to a producers warehouse Inventory quantity on hold of raw materials from time 0 to tPW Receiving cost of raw materials per receiving for a producer (constant) Holding cost of raw materials per unit per unit time for a producers warehouse (constant) Cost of a deteriorated unit of raw material for a producer (constant) Total cost for a producers warehouse Production period Maximum inventory level of finished goods for a producer during period t P Inventory quantity on hold of finished goods from time 0 to t p Set-up cost for a producer per set-up (constant) Delivery cost of finished goods per delivery for a producer (constant) Holding cost of finished goods per unit per unit time for a producer (constant) Cost of a deteriorated unit for a producers finished goods (constant) Total cost for a producer Maximum inventory level of finished goods per receiving for a buyer Ordering cost of finished goods for a buyer (constant) Receiving cost of finished goods for a buyer (constant) Holding cost of finished goods per unit per unit time for a buyer (constant) Cost of a deteriorated unit for a buyers finished goods (constant) Total cost for a buyer

43

3.

The study of Rau et al. (2004)

44

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

Rau et al. (2004) developed an integrated inventory model of a supply chain for deteriorating items among a supplier, a producer and a buyer. In his study, he discussed about two inventory model cases; the first is without shortage, and the second is with shortage. However, in our study, we will only discuss the inventory model without shortage for each partner in the supply chain. We address the whole equations which includes inventory model of a buyers finished goods, inventory model of a producers finished goods, inventory model of a producers raw materials, inventory model of suppliers raw materials, and integrated inventory model as follows. 3.1 Inventory model of a buyers finished goods The finished goods inventory level at the extreme point:

and the inventory quantity of finished goods for the buyer from time 0 to t B is:

The total cost for finished goods for the buyer per T can be expressed as the sum of the order cost, receiving cost, holding cost, deterioration cost, i.e.

3.2 Inventory model of a producers finished goods Maximum inventory level of finished goods for a producer during period t P as:

Inventory quantity on hold of finished goods for a producer from time 0 to tp is:

The total finished goods cost for the producer per T can be expressed as the sum of the set-up cost, delivery cost, holding cost, deterioration cost, as follows:

3.3 Inventory model of a producers raw materials Quantity of raw materials per delivery from a supplier to a producers warehouse can be expressed using the following equation:

Inventory quantity on hold of raw materials from time 0 to tP is:

The total cost for the producers raw materials wareho use per T can be expressed as the sum of the receiving cost, holding cost, and deteriorating cost, as follows:

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

45

3.4 Inventory model of suppliers raw materials Total quantity of raw materials receiving (planned) from a suppliers vendor during period T is:

qs

q pw (1Ts )t

(10)

Inventory quantity on hold of raw materials for supplier from time 0 to t is:

The total raw materials cost for the supplier per T can be expressed as the sum of the order cost, delivery cost, holding cost, deterioration cost, as follows:

3.5Integrated inventory model The integrated total cost for the buyer, producer and supplier, TC, as follows:

46

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

4.

System dynamics model

System dynamics (SD) modeling and simulation was designed for modeling and analysis of complex socio-economic systems by Jay W. Forrester (1961) at the Massachusetts Institute of Technology. Since then, SD has been applied to help business policy making and solving dynamic industrial management problems (Forrester, 1961; Senge, 2006; Erma et al., 2010). Some related studies are as follows, sterman (2000) discusses case studies about modeling logistics problems by SD methodology. Minegishi and Thiel (2000) reveal the complex logistic behavior of a food industry by applying SD. Their study sheds light on coordinating of variables in the food production. Ozbayrak, M. et al. (2007) developed a systems dynamics model of a manufacturing supply chain system to model the operation of the supply chain network under study and obtain a true reflection of its behavior. This methodology can be used in the simulation of complicated, nonlinear, multi-loop feedback systems by developing the causal and feedback loop. In this paper, we applied Vensim (Ventana Systems Inc., 2004) as the system dynamics modeling and simulation software. Vensim is a professional simulation software, and it is a graphical user interface. We can conduct a system on the basis of the causal relationships and feedback loops. There are four types of items in a system dynamics model: (1) flow, (2) stock, (3) converter (or auxiliary) and (4) connectors. Flow variables which either fill in or consume the stock represent the rate of change in the stock variables. Stock variables are the major accumulations in the system, and they are the state variables. Converters are intermediate variables for calculations, and are represented by general variables. Finally, the connectors which represent the cause-and-effect directions are represented by simple arrows. 4.1 Causal Loop Diagram According to the logic and function of integrated inventory model of a supply chain for deteriorating items, we developed the causal loop diagram as figure 1. The diagram is consisted of a buyer, a supplier, and a producer which has to be considered by owning different inventories of raw material and finished goods. As showed in figure 1, we can realize the following logics. When a buyers demand increases, a producers finished goods on hand decreases. When a producers finished goods on hand increases, the producers finished goods inventory increases. When a producers order 2 increases, the producers finished goods inventory increases and the producers raw material inventory decreases. When a producers order 1 increases, the producers raw material inventory increases and the suppliers total inventory decreases. When a suppliers order increases, the suppliers total inventory increases. According to figure 1, we developed a system dynamics simulation diagram as figure 2. Then, we added order signal into that and proposed a new order system as figure 3. < Fig. 1 > < Fig. 2 > < Fig. 3 > After developing the cause and effect loop diagram of the integrated supply chain model by Vensim, we need to discuss about the influence level of objectives by inputting different significant decision making attributes and factors. First, we need to input the variables values as an example which used here is an extension of Rau et al. (2004)s. The variable values are as follows:

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

47

z z

Planning cycle T: a unit time (example: 1 week). Buyers parameters data: demand rate (D)=12000 units per week; ordering cost (AB)=$300 per order; receiving cost(FB)=$25 per receiving; holding cost (HB)= $15 per order; deterioration cost (PB)= $110 per unit; deterioration rate ( qB )= 0.08. Producers parameters data: production rate (P)= 24000 units per week; set-up cost (SP)= $500 per setup; delivery cost (FP)=$150 per delivery; holding cost (HP)= $12 per unit per week; deterioration cost (PP)= $90 per unit; deterioration rate of finished goods( qP )= 0.095; receiving cost of raw material(FPw)= $20 per receiving; holding cost of raw material (H PW)=$10 per unit per week; deterioration cost (P PW )=$85 per unit; deterioration rate of raw material( qPW )= 0.09; raw materials reorder point (PW_order_ point) = 50. Suppliers parameters data: ordering cost (AS)= $250 per order; receiving cost (F S)=$125 per receiving; holding cost (HS)= $8 per unit per week; deterioration cost (P S)= $75 per unit; deterioration rate (qs)= 0.1; reorder point (S_order_ point)=150.

48

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

4.2 The proposed new order system Policy 1: Simulating the order system of Rau et al. (2004), the flow chart as figure 4. Policy 2: Modifying the order system of Rau et al. (2004) to be a new proposed order system, the flow chart as figure 5. < Fig. 4 > < Fig. 5 > The setting and action logics of each decision attributes and the function of time of casual factors in the proposed order system are described as follows. (a) suppliers order law V_shipping = qs S_order_signal Among which calculation logic of S_order_signal is as follow: If S_ inventory S_order_point, Then S_order_signal = 1, else S_order_signal =0 (b) producers raw material order law S_shipping = qPW PW_order_signal Among which calculation logic of PW_order_signal is as follow: If PW_inventory PW_order_ point, Then PW_order_signal = 1, else PW_order_signal =0 4.3 Validation and Model testing After modeling dynamic systems, to reflect the models practicability and validity, model testing is a must. Model testing can increase the decision maker s confidence to models validation and then can provide the correct decision making for corporate. The items about validation of model testing includes structure assessment, parameter assessment, extreme conditions, boundary adequacy, dimensional consistency, behavior anomaly, behavior reproduction, and sensitivity analysisSterman, 2000. We then conducted model testing according to those items separately and the results showed that our model is appropriate. 5. Model analysis and numerical example for applications

We compared the order systems of Rau et al. (2004) and the proposed new one in this study. The results showed: under the simulation time of 1 week, the model of Rau et al. (2004) and the proposed one are consistent and appear to be divergent. The total cost at week one of Rau et al. (2004) and the proposed order system separately are 35,394 dollars and 33,566 dollars. The delivery numbers are both 23. From the comparison, we know that the proposed new order system has lower total cost. 6. Conclusion

In this paper, we make an extension from the study of Rau et al. (2004). We apply system dynamics simulation methodology to conduct a systematically simulation for an inventory model for deteriorating items in a supply chain. The results are as follows: we developed a dynamic system model to analyze inventory policy and the dynamic relations among inventory costs. The customers may change their demands constantly, and the suppliers and producers have to modify their inventory model frequently to effectively lower the cost of each others. This paper modified the order system proposed by Rau et al. (2004), and suggested that the system dynamic model and simulation can provide a more thorough, robust and long-term perspective solution. The simulation results showed that system dynamics simulation methodology is proper to be applied at time evolution related research subjects. Acknowledgments This study is supported by the National Science Council of Taiwan, Republic of China, under grant number NSC 99-2221-E-451-010-MY2.

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

49

References
[1] Alamri, A. A. and Balkhi, Z. T. (2007). The effects of learning and forgetting on the optimal production lot size for deteriorating items with time varying demand and deterioraion rates, International Journal of Production Economics, 107, 125-138. [2] Chorpa, S. and Menidl, P. (2001). Supply chain management: Strategy, planning and operations, Prentice-Hall. [3] Erma, S., Shuo, Y. C., Rudi, H. and Chih, H. C. (2010). Demand scenario analysis and planned capacity expansion: A system dynamics framework, Simulation Modelling Practice and Theory, 18(6), 732-751. [4] Forrester , W . (1961). Industrial Dynamics, Cambridge, Mass: The M. I. T. Press. [5] Ghare, P. M. and Schrader, G. F. (1963). A model for an exponentially decaying inventory, Journal of Industrial Engineering, 14, 238-243. [6] Hsu, P. H., Wee, H. M. and Teng, H. M. (2010). Preservation technology investment for deteriorating inventory, International Journal of Production Economics, 124, 388-394. [7] Krajewski, L. J., Ritzman, L. P. and Malhotra, M. K. (2010). Operations management: Processes and supply chains, Upper Saddle River, NJ: Pearson Prentice Hall. [8] Kung, J. W. Lin, Y. S. and Jonas C. P. Y. (2010). Optimizing inventory policy for products with time-sensitive deteriorating rates in a multi-echelon supply chain, International Journal of Production Economics, 1-11. [9] Minegishi, S., and Thiel, D. (2000). System dynamics modeling and simulation of a particular food supply chain, Simulation Practice and Theory, 8, 321-339. [10] Minner, S. (2003). Multiple-supplier inventory models in SCM: A review, International Journal of Production Economics, 81/82, 265-279. [11] Ozbayrak M., Papadopoulou T. C., and Akgun M., (2007). Systems dynamics modeling of a manufacturing supply chain system, Simulation Modeling Practice and Theory, 15, 1338-1355. [12] Ping, H. H., Hui, M. W. and Hui, M. T. (2007). Optimal ordering decision for deteriorating items with expiration date and uncertain lead time, Computers and Industrial Engineering, 52, 448-458. [13] Ranjan, B. and Susmita, B. (2011). A review of the causes of bullwhip effect in a supply chain, The International Journal of Advanced Manufacturing Technology, 54, 1245-1261. [14] Rau, H., Wu, M. Y. and Wee, H. M. (2004). Deterioration item inventory model with shortage due to supplier in an integrated supply chain, International Journal of System Science, 35, 293-303. [15] Senge, P. M. (2006). The fifth discipline: the art and practice of the learning organization , New York: Doubleday Currency. [16] Sterman, J. (2000). Business dynamics: Systems thinking and modeling for a complex world, Boston: Irwin McGraw-Hill. [17] Ventana Systems Inc. (2004), Vensim Version 5 Users Guide, Harvard: Ventana Systems Inc. Press. [18] Wee, H. M. (1999). Deteriorating inventory model with quantity discount, pricing and partial backordering, International Journal of Production Economics, 59, 511-518. [19] Yi , H. L., Chinho, L. and Binshan, L. (2010). On conflict and cooperation in a two-echelon inventory model for deteriorating items, Computers and Industrial Engineering, 59, 703-711. [20] Yong, H., Shou Y. W. and Lai, K. K. (2010). An optimal production-inventory model for deteriorating items with multiple-market demand, European Journal of Operational Research, 203, 593-600.

Figures and Tables

The supplier's total inventoy + a supplier's order

The producer's raw material inventory a producer's order 2

The producer's finished goods inventory + a producer's finished goods on hand

a buyer's demand

a producer's order 1

Fig. 1 Casual loop diagram

50

V shipping PW shipping P shipping

S inventory

P warehouse

P inventory

V shipping S shipping PW shipping PW order signal qP Det PW PW order point Det B S order signal

S inventory

P warehouse

P inventory P shipping

S shipping

qS qP Det B

qS <tB> S order point Det S

Det PW

<tB>

Det S Det P qB Wp

q PW

q PW P Ws

Det P

Wp qB

Ws

<n> tP D <tB>

<tP>

<n> <T> Hs
<WB> n

<tP> tP <tB> D <WB>

<T>

TCs

TCs As Ps <Wpw>

Hs

Fs <tB> TCB HB

As Ps

Fs Fpw Hpw TC Sp

<Wpw> AB

n TCpw <tB> Ppw PB FB Pp <TCp> Hp T Fp TCB HB AB

Fpw PB FB T

TCpw

Ppw

Hpw

TC

Sp

Pp

Fp

<TCp>

Hp

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

Fig. 2 System dynamics simulation diagram

Fig. 3 System dynamics simulation diagram of proposed new order system

Ching-Fang Lee and Chien-Ping Chung / Procedia - Social and Behavioral Sciences 40 (2012) 41 51

51

Inventory available New Demands

Order screening period?

N
Reject order

Accept order

Send order
Fig. 4 The order system of Rau et al. (2004)

Inventory available New Demands

Order Screening period?

N
Reject order

Y
Inventory screening

Inventory order

Y
Accept Order Send order
Fig. 5 Proposed new order system