Journal Search Engine

ISSN : 2005-0461(Print)
ISSN : 2287-7975(Online)

Journal of Society of Korea Industrial and Systems Engineering Vol.37 No.4 pp.231-238
DOI : https://doi.org/10.11627/jkise.2014.37.4.231

A Study of Inventory Models for Imperfect Manufacturing Setup Considering Work-in-Process Inventory

Misbah Ullah, Chang W. Kang†, Shehereyar Mohsin Qureshi

Department of Industrial and Management Engineering, ERICA Campus, Hanyang University

Corresponding Author : cwkang57@hanyang.ac.kr

Received November 10, 2014 Finally Revised December 6, 2014 Accepted December 6, 2014

Abstract

Key Words :

재공품 재고를 고려한 제조 시스템에서의 재고 관리 모델 연구

미스바 울라, 강 창욱†, 쉐리야르 모신 코레시

한양대학교 산업경영공학과

초록

Optimum lot size calculation for real world manufacturing environment has been focused since last few decades. Several extensions have been made to the basic economic order and production order quantity models to realize the possible practical situations in industry. However, focus on work-in-process inventory has been ignored relatively. This paper provides a comprehensive review of the models developed for group technology based manufacturing environment focusing on work-in-process inventory. Models have been extended from a perfect manufacturing conditions to an imperfect manufacturing situation considering rework, rejection and inspection. Optimum lot size has been evaluated using a simple algebraic optimization approach. Significant parameters are highlighted using sensitivity analysis for the developed models. Numerical example is used to illustrate the utilization of such models in day-to-day production setups and the impact of significant factors’ variation on total cost and optimum lot size.

키워드 : Work-In-Process , Imperfect , Lot Size , Economic Order Quantity

This article has been cited by 0 article in crossref

Cited-By

Funding:

1.Introduction

Inventory models have been widely used in todays’ manufacturing and service industries in order to obtain optimum lot size with average cost minimization. Harris [1] introduced economic order quantity model for the first time in Feb, 1913. The model was later extended to production order quantity model in 1918. Since then, an extensive research has been carried out to develop models for manufacturing environments keeping average cost minimization and demands fulfillment on priority.

One major reason for these extensions is the assumption of the aforementioned models regarding processes perfection. Economic and production order quantity models assume perfect product quality. As processes are affected by a number of factors including tool wear and tear, material internal defects, machines failure and breakdowns, and environmental conditions etc. Therefore several extensions to these basic models have been made during past few decades. Imperfect production units can be reworked and its ultimate impact on lot size was initially studied by Gupta and Chakraborti [2]. Lee and Rosenblatt [3] gave the idea of increased investment in quality control techniques. They concluded that this will help organization to reduce defect rate and other inventory cost including setup cost and carrying cost. Cheng [4] focused imperfect production setup and developed a model for economical quantity order based on demand dependency for a unit production cost. Ben-Daya and Harija [5] assumed that deviation of production process from an ideal conditions is a random process. These deviations results nonconforming products. Economic order quantity model was extended for electronic units manufacturing by Salameh and Jaber [6]. They assumed that all incoming products are screened before they are processed in a manufacturing setup. It was further added that non-reworkable products are sold in a single batch at low price. Goyal and Cardenas-Barron [7] introduced simple algebraic optimization technique to calculate optimum economic order quantity. The results obtained via this approach were similar as obtained through other methods of optimization.

Biswas and Sarker [8] introduced the idea of lean manufacturing into the lot sizing inventory problems. They assumed that scrap items are recognized ‘before’, ‘during’ and ‘after’ the rework operation. Buffer stations idea was also incorporated to meet the customer demands and avoid shortages. Jamal et al. [9] developed model based on the assumption that reworking process can be executed either within the cycle or at the end of all cycles. Other who added in this particular direction include are [10-17, 26, 27]. Wee et al. [18] calculated lot size based on total average profit in order to optimize economic order quantity. They used renewal reward theorem (RRT) in addition to screening constraints and imperfect manufacturing processes. Researchers established that reworking an imperfect items is of significance for the manufacturing company. [19, 20, 21]. Khan et al. [22] presented a brief review of economic order quantity model for imperfect items. They mainly focused model developed by Salameh and Jaber. Their review categorized different extended models based on quality, imperfect items produced, shortages and backordering, and supply chain perspective etc.

Models considered above mainly focused two kinds of inventories i.e. raw material inventory and finish goods inventory. However, the third kind of inventory i.e. work-in-process (WIP) inventory has been ignored relatively. The role of work-in-process inventory in group technology work environment, in particular, has been significant. Lot sizing models are required to incorporate work-in-process inventory cost while optimizing the total cost function. The role of work-in -process inventory can be more substantial when the manufacturing processes are imperfect. This paper has presented a brief re-view of models developed for optimum lot size calculation focusing work-in-process inventory. Paper highlight significant parameters in each model with numerical examples using sensitivity analysis. The remaining sections of this paper are ordered as follows. Next section gives an overview of the inventory models developed for work-inprocess inventory with assumptions and notation used. Section 3 gives illustration of the developed models through numerical examples in addition to sensitivity analysis. Conclusion and possible future extensions have been discussed in the last section.

2.Group Technology Order Quantity Model

Boucher [23] introduced the concept of work-in-process inventory for group technology manufacturing environment. The traditional economic order quantity model was extended by incorporating work-in-process cost into the total cost function with these assumptions. 1) Shortages are not allowed. 2). Demand is known and constant. 3). Manufacturing processes are perfect. 4). Inspection takes no time. 5). All other parameters including demand rate, set up and manufacturing time etc. are fixed and pre-determined.

Following notation have been used to develop the model.

Parameters

D demand rate of good quality product (unit product per unit time)

s setup time per cycle (unit time per setup)

m machining time per unit for lot size Q (unit time per unit product)

T_c cycle time

T_p total processing time

T average manufacturing time for each product item

C_M raw material cost per unit ($/unit of product)

C_p cost of purchase per unit of time ($/unit of time)

C_s setup cost per unit of time ($/unit of time)

C_wip work in process holding cost per unit of time ($/unit of time)

C_H carrying cost per unit of time ($/unit of time)

C_total cost (total) per unit of time ($/unit of time)

I average storage inventory

W average monetary value of the WIP inventory ($)

i inventory holding cost per unit of time ($/unit of time)

c average unit value of each product cost (unit of money ($) per unit)

R rate charged per unit of cell production time including all overheads, moving cost, loading/unloading cost, etc. (unit of money ($) per unit of time)

Decision Variable

Q production lot size per cycle (number of product units per cycle)

The total cost function include setup cost, finish components carrying cost and work-in-process (WIP) inventory carrying cost. Work-in-process carrying cost is associated with products carried during its manufacturing phase till their production.

GTOQ model, total cost function is given by

\begin{matrix} C_{total} = \frac{AD}{Q} + \frac{iQ}{2} (C_{M} + \frac{s}{Q} R + mR) \\ + iD (C_{M} + \frac{s}{2Q} R + \frac{m}{2} R) (s + mQ) \end{matrix}

(1)

where A is the setup cost and is equal to the setup time (s) multiplied by rate charged per unit of the cell production (R).

Algebraic optimization method can be used to obtain the optimum lot size [24] without using differentiation method of optimization.

Equation (1) can be rearranged as,

\begin{matrix} C_{total} = \frac{AD}{Q} + \frac{iQ C_{M}}{2} + \frac{isR}{2} + \frac{iQmR}{2} + i D_{s} C_{M} \\ + \frac{iD s^{2} R}{2 Q} + iDmsR + iDmQ C_{M} + \frac{iD m^{2} RQ}{2} \end{matrix}

Factorizing above Equation,

\begin{matrix} C_{total} = (\frac{i C_{M}}{2} + \frac{imR}{2} + iDm C_{M} + \frac{iD m^{2} R}{2}) \\ Q + (AD + \frac{iD s^{2} R}{2}) \frac{1}{Q} + \frac{isR}{2} + iDs C_{M} + iDmsR\} \end{matrix}

(2)

Equation (2) is of the form $y (x) = a_{1} x + \frac{a_{2}}{x} + a_{3}$ . Therefore using algebraic optimization method to obtain the optimum lot size as follow,

\begin{matrix} Q^{*} = x = \sqrt{\frac{a_{2}}{a_{1}}} = \sqrt{\frac{AD + \frac{iD s^{2} R}{2}}{\frac{i C_{M}}{2} + \frac{imR}{2} + iDm C_{M} + \frac{iD m^{2} R}{2}}} \\ Q^{*} = \sqrt{\frac{(AD + \frac{iD s^{2} R}{2})}{i (\frac{C_{M} + mR}{2}) + Dmi (C_{M} + \frac{mR}{2})}} \end{matrix}

(3)

2.1.Group Technology Order Quantity Model Considering Rework (GTOQR)

The GTOQ model focused perfect production processes and it was assumed that no rework or rejection will be produced by the manufacturing cell. However, in practical situations, the concept of rework is common especially in manufacturing cells. Barzoki et al. [25] extended the GTOQ model by incorporating the rework operation into the manufacturing lot focusing work-in-process inventory. This new extended model, known as group technology order quantity model with rework consideration (GTOQR), assumed that process may produce imperfect products which are either reworkable or non-reworkable. The demand for non-reworkable products always exist and can be sold out at reduced price. Few assumptions were made in order to derive this model in addition to GTOQ model assumptions. 1). No imperfect products are produced during the rework phase 2). Inspection takes no time. 3). Process can be stopped for rework operation at any stage of manufacturing a lot. 4). Rework-able and non-reworkable imperfect products’ proportion is known and constant.

Several notation are added to develop the model.

Parameters

p₁ reworkable products percentage (%)

p₂ non-reworkable products percentage (%)

C_p purchase cost per unit of time ($/unit of time)

The model also added material purchase cost and inspection cost to calculate the total cost function. Therefore, total cost function comprises purchase cost, setup cost, inspection cost, work-in-process carrying cost and inventory holding cost for finish goods.

Therefore,

$C_{total} = \frac{{DC}_{M}}{(1 - p_{2})} + \frac{AD}{Q (1 - p_{2})} + \frac{ID}{(1 - p_{2})} + \frac{i}{2} ((\frac{D}{1 - p_{2}}) (s + mQ (1 + p_{1})) (2 C_{M} + \frac{Rs}{Q} + mR (1 + p_{1}))) + \frac{i}{2} (1 - p_{2}) Q (C_{M} + \frac{Rs}{Q} + mR (1 + p_{1}))$

Factorizing above Equation

\begin{matrix} \begin{matrix} C_{total} = (\frac{i (1 - p_{2})}{2} (C_{M} + mR (1 + p_{1})) \\ + \frac{Dim}{2 (1 - p_{2})} (2 C_{M} + mr (1 + p_{1})) \end{matrix} \\ Q + (\frac{AD}{(1 - p_{2})} + \frac{{Dis}^{2} R}{2 (1 - p_{2})}) \frac{1}{Q} + \frac{D}{(1 - p_{2})} \\ (C_{M} + I + i (C_{Ms} + R_{ms} + R_{{msp}_{1}}) + \frac{iRs}{2} (1 - p_{2}) \end{matrix}

(4)

Equation (4) is of the form $y (x) = a_{1} x + \frac{a_{2}}{x} + a_{3}$ . Therefore, using algebraic optimization method to obtain the optimum lot size as follow :

$\begin{matrix} Q^{*} = x = \sqrt{\frac{a_{2}}{a_{2}}} \\ \sqrt{\frac{\frac{AD}{(1 - p_{2})} + \frac{{Dis}^{2} R}{2 (1 - p_{2})}}{\frac{i (1 - p_{2})}{2} (C_{M} + mR (1 + p_{1})) + \frac{Dim}{2 (1 - p_{2})} (2 C_{M} + mr (1 + p_{1}))}} \end{matrix}$

Rearranging above Equation

Q^{*} = \sqrt{\frac{2 AD + {Dis}^{2} R}{i {(1 - p_{2})}^{2} (C_{M} + mR (1 + p_{1})) + Dim (2 C_{M} + mr (1 + p_{1})}}

(5)

The model highlighted the impact of reworkable and nonreworkable products produced during the manufacturing process. The model is considered as extension of the basic GTOQ model as the optimum lot size calculated via GTOQR model is the same if we consider the manufacturing process as perfect. The resultant lot size will be the same as the one calculated by GTOQ model. The proof can be obtained easily if we assume that imperfection doesn’t exist in manufacturing process i.e. p₁ = p₂ =0

2.2.Group Technology Order Quantity Model Considering Rework and Inspection

The GTOQR is based on several assumptions including no inspection time, rework operation is a perfect process, and reworkable products qualification without inspection etc. Ullah and Kang [26] extended this model by incorporating 1) inspection time into the production cycle. 2) Lot manufacturing was divided into two phases, in second phase only reworkable products were processed and they were classified into either accepted or rejected. 3) Inspection for reworkable products is assumed to be mandatory as they were relatively more sensitive regarding qualification. 4) Previous WIP based GTOQR model assumed that manufacturing station can be stopped at any time for rework operation. This assumption was considered against the general practice in industry as usually rework is done when complete lot is passed through the manufacturing process. This extended model was named as group technology order quantity model considering inspection and rework (GTOQIR).

This model assume that machining process is followed by an inspection process which consume time and resources. Inspection process results in either rejected units, reworkable units, or accepted products as shown in <Figure 1>. Reworkable products are processed again through the machining process followed by an inspection process. Inspection station classify unit products as either qualified or rejected as shown in <Figure 2>.

Parameters

m₁ machining time per unit for the lot size Q in phase 1. (unit time per unit product)

m_r machining time per unit for rework products (unit time per unit product)

P₁ percentage of products rejected as poor quality in phase 1

P₁₁ percentage of products qualified as good quality in phase 1

P₂₁ percentage of products qualified as good quality in phase 2

P₂₀ percentage of products rejected as poor quality in phase 2

P_1r percentage of products rework able in phase 1, QP_1r = Q(PP₂₀+P₂₁)

P_b percentage of poor quality products at the end of cycle

P_g percentage of good quality products at the end of cycle

The total cost function include purchase cost, setup cost, inspection cost, work-in-process inventory carrying cost and finish goods inventory holding cost.

Therefore, total cost function per unit time is given by

C_{total} = C_{p} + C_{s} + C_{i} + C_{H} + C_{wip}

(6)

Ullah and Kang [25] developed mathematical expressions for the above mentioned costs. Purchase cost is calculated for raw materials purchased in each cycle. Setups cost is calculated only once for lot size Q during each cycle.

been used. Equation (8) is of the WIP composed of carrying cost for products waiting for operation, cost of carrying rejected products, and cost of carrying good quality products during the cycle time. Inspection cost is calculated for lot manufacturing during regular time and for reworked products. Therefore, the total cost function is,

\begin{matrix} C_{total} = \frac{C_{M} D}{(1 - P_{b})} + \frac{AD}{Q \{1 - P_{b}\}} + \{\frac{1 D}{(1 - P_{b})}\} (1 + P_{1 r})\} \\ + \frac{1}{2} i (\{C_{M} + R (\frac{s + {Qm}_{1} + IQ + {Qm}_{r} P_{1 r} + IQ P_{1 r}}{Q})\}) \\ + Q (1 - P_{b}) + \frac{1}{2} [\frac{Di}{1 - P_{b}}] (2 C_{M} + \frac{Rs}{Q} + {Rm}_{1} + {Rm}_{r} P_{1 r} \\ + IR + {IRP}_{1 r}) \times (s + {Qm}_{1} + Q P_{1 r} m_{r} + QI + {QIP}_{1 r}) \end{matrix}

(7)

Assuming

$\begin{matrix} K_{1} = ((C_{M} + {Rm}_{1} + IR) + ({Rm}_{r} + IR) P_{1 r}) \\ K_{2} = ((m_{1} + m_{r} P_{1 r}) (2 C_{M} + {Rm}_{1} + {Rm}_{r} P_{1 r})) \\ K_{3} = (I (1 + P_{1 r}) (2 C_{M} + IR + 2 {Rm}_{1} + {2Rm}_{r} P_{1 r} + IR P_{1 r})) \end{matrix}$

then, the total cost function can be rearranged in factorized form as follows :

\begin{matrix} C_{total} = (\frac{i}{2} (1 - P_{b}) K_{1} + (\frac{Di}{2 (1 - P_{b})}) (K_{2} + K_{3})) Q + \\ (\frac{Di}{2 (1 - P_{b})} ({Rs}^{2}) + \frac{AD}{(1 - P_{b})}) \frac{1}{Q} + (\frac{C_{M} D}{(1 - P_{b})} + (\frac{ID}{(1 - P_{b})}) \\ (1 + P_{1 r}) + \frac{Di}{2 (1 - P_{b})} \\ ({Rsm}_{1} + {Rsm}_{r} P_{1 r} + RsI + {RsIP}_{1 r}) \end{matrix}

(8)

In order to obtain the optimum lot size, algebraic optimization technique has been used. Equation (8) is of the form of the Equation $y (x) = a_{1} x + \frac{a_{2}}{x} + a_{3}$ , therefore, following Equation is obtained for the optimum lot size Q

$\begin{matrix} Q^{*} = x = \sqrt{\frac{a_{2}}{a_{1}}} \\ = \sqrt{\frac{\frac{AD}{(1 - P_{b})} + \frac{Di}{2 (1 - P_{b})} ({Rs}^{2})}{(\frac{i}{2} (1 - P_{b}) K_{1} + (\frac{Di}{2 (1 - P_{b})}) (K_{2} + K_{3}))}} \end{matrix}$

Rearranging,

Q^{*} = \sqrt{\frac{2 AD + iDR {(s)}^{2}}{i {(1 - P_{b})}^{2} K_{1} + Di (K_{2} + K_{3})}}

(9)

Equation (9) give the closed form optimum solution for the lot size (Q).

3.Numerical Computation and Sensitivity Analysis

For sensitivity analysis all above mentioned models are analyzed using following example. Different parameters of the aforementioned models are varied from -50% to +50% to realize the changes in the parameter value and its ultimate effect on the optimum lot size. The results are summarized in tabular form.

A numerical example from GTOQR model [24] has been used for this purpose. The data assume include A =11.9 ($/unit), D = 14000 (units/year), i = 35%, C_M = 1($/unit), s = 0.0017(years/unit), m₁ = 0.12 (mints/unit), m_r = 5% (m₁), R = 7000 ($/year). It is further added that rework rate has been taken as 5% of the regular production time and rejection rate as 20% during calculation.

Six different parameters including setup cost, demand rate, manufacturing time, rework percentage, rejection percentage, and inspection time have been analyzed against each model and their responses have been shown in following Tables.

Group Technology Order Quantity (GTOQ) and its onward extensions till date (GTOQR and GTOQIR models) analysis has been shown in <Table 1>, <Table 2> and <Table 3> respectively. It can be observed that the change in lot size with respect to setup cost variation is significant. Lot size increase with increases in setup cost and vice versa. GTOQ Model results in lowest lot size value (Q* = 678) whereas GTOQR model response to the setup cost produce maximum value (Q* = 1455) under the given conditions. GTOQR and GTOQIR value is increased due to the processes imperfection.

Demand rate has also been varied from -50% to +50% against the optimum lot size and variation can be observed for all the three models. It is shown that demand rate effect on lot size is also considerable. Lot size increases with increase in demand rate per year. Higher demand rate results in higher values of lot size for GTOQR model. The values of GTOQIR are relatively reduced due to the concept of rework operation involved.

Manufacturing time affect cycle time. Manufacturing time increasing affect results in decreasing the lot size in a manufacturing cell. As time for manufacturing decreases, lot size for group technology manufacturing environment increases. The GTOQR model give highest value (Q* = 1204) in comparison to Q* = 967 and Q* = 1174 for GTOQ and GTOQIR models respectively.

Rework operation on optimum lot for various values of rework percentage has been shown in all the three Tables. There is no change in the optimum lot size for GTOQ model as it doesn’t take into consideration processes imperfection. GTOQR and GTOQIR model values are changed less significantly when rework operation is varied from 2.5% to 7.5% of the lot size. This impact can be significant for higher values of rework percentage in both of these models.

Change in rejection percentage in a cycle time and its impact on lot size can be observed in GTOQR and GTOQIR models. GTOQ model don’t consider rejection in manufactu ring processes. The lot size increases as the rejection percentage increases from 10% to 30%. Optimum lot size calculated by GTOQIR model is relatively low as compared to GTOQR model. The change in lot size for both the models is significant as shown in <Table 2> and <Table 3> respectively.

At the end the impact of inspection time required for unit product qualification can be analyzed. It can be observed that optimum lot size for GTOQ and GTOQR model remains unchanged for different values of inspection time as shown in <Table 1> and <Table 2>. However this effect can be observed in <Table 3> for GTOQIR model. Lot size has been reduced for higher values of inspection time per unit product.

4.Conclusion

The paper summarize inventory models developed for group technology work environment focusing work-in-process inventory. Models are derived in a chronological order i.e. from a perfect manufacturing setup to an imperfect manufacturing situation considering rework, rejection and inspection process.A simple algebraic optimization approach has been used to get the optimum lot size for the developedc models. Mathematical models for optimum lot size under different working conditions have been illustrated with numerical example. Sensitivity analysis for several parameters has been highlighted via numerical examples. The impact of demand rate, setup cost, manufacturing time, rejection and inspection time remain significant for calculation of optimum lot size in an imperfect manufacturing setup. The paper provide an insight to manufacturing engineers to calculate optimum lot size under different imperfect manufacturing conditions with average cost function minimization. Further extension are possible using probabilistic approach for imperfect manufacturing setup focusing work-in-process inventories.

Figure

<Figure 1>.

Phase 1

<Figure 2>.

Phase 2

Table

<Table 1>.

Sensitivity Analysis (GTOQ Model)

S N	Parameter	Changes(%)	Q*
01	Setup cost (A)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>678</p><p>831</p><p>1072</p><p>1175</p>
02	Demand rate (D)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>683</p><p>833</p><p>1069</p><p>1167</p>
03	Manufacturing time (m1)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>967</p><p>963</p><p>955</p><p>951</p>
04	Rework (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>959</p><p>959</p><p>959</p><p>959</p>
05	Rejection (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>959</p><p>959</p><p>959</p><p>959</p>
06	Inspection time (I)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>959</p><p>959</p><p>959</p><p>959</p>

<Table 2>.

Sensitivity Analysis (GTOQR Model)

S N	Parameter	Changes(%)	Q*
01	Setup cost (A)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>840</p><p>1029</p><p>1329</p><p>1455</p>
02	Demand rate (D)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>850</p><p>1035</p><p>1322</p><p>1440</p>
03	Manufacturing time (m1)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1204</p><p>1196</p><p>1181</p><p>1174</p>
04	Rework (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1189</p><p>1189</p><p>1188</p><p>1188</p>
05	Rejection (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1061</p><p>1121</p><p>1264</p><p>1349</p>
06	Inspection time (I)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1188</p><p>1188</p><p>1188</p><p>1188</p>

<Table 3>.

Sensitivity Analysis (GTOQIR Model)

S N	Parameter	Changes(%)	Q*
01	Setup cost (A)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>821</p><p>1005</p><p>1297</p><p>1421</p>
02	Demand rate (D)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>838</p><p>1015</p><p>1284</p><p>1393</p>
03	Manufacturing time (m1)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1174</p><p>1167</p><p>1154</p><p>1147</p>
04	Rework (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1161</p><p>1161</p><p>1160</p><p>1160</p>
05	Rejection (%)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1041</p><p>1097</p><p>1231</p><p>1310</p>
06	Inspection time (I)	<p>-50%</p><p>-25%</p><p>+25%</p><p>+50%</p>	<p>1175</p><p>1168</p><p>1153</p><p>1147</p>

Reference

Harris FW (1990) How many parts to make at once , Operations Research, Vol.38 (6 ) ; pp.947-950
Gupta T , Chakraborty S (1984) Looping in a multistage production system , The International Journal of Production Research, Vol.22 (2 ) ; pp.299-311
Rosenblatt MJ , Lee HL (1986) Economic production cycles with imperfect production processes , IIE transactions, Vol.18 (1 ) ; pp.48-55
Cheng T (1991) An economic order quantity model with demand- dependent unit production cost and imperfect production processes , IIE transactions, Vol.23 (1 ) ; pp.23-28
Ben-Daya M , Hariga M (2000) Economic lot scheduling problem with imperfect production processes , Journal of the Operational Research Society, Vol.51 (7 ) ; pp.875-881
Salameh M , Jaber M (2000) Economic production quantity model for items with imperfect quality , International Journal of Production Economics, Vol.64 (1 ) ; pp.59-64
Goyal SK , Cardenas-Barron LE (2002) Note on: economic production quantity model for items with imperfect quality -a practical approach , International Journal of Production Economics, Vol.77 (1 ) ; pp.85- 87
Biswas P , Sarker BR (2008) Optimal batch quantity models for a lean production system with in-cycle rework and scrap , International Journal of Production Research, Vol.46 (23 ) ; pp.6585-6610
Jamal A , Sarker BR , Mondal S (2004) Optimal manufacturing batch size with rework process at a single-stage production system , Computers and Industrial Engineering, Vol.47 (1 ) ; pp.77-89
Chiu SW , Ting C-K , Chiu Y-SP (2007) Optimal production lot sizing with rework, scrap rate, and service level constraint , Mathematical and Computer Modelling, Vol.46 (3-4 ) ; pp.535-549
Chan W , Ibrahim R , Lochert P (2003) A new EPQ model : integrating lower pricing, rework and reject situations , Production Planning and Control, Vol.14 (7 ) ; pp.588-595
Pal B , Sana SS , Chaudhuri K (2013) Maximising profits for an EPQ model with unreliable machine and rework of random defective items , International Journal of Systems Science, Vol.44 (3 ) ; pp.582-594
Ojha D , Sarker B , Biswas P (2007) An optimal batch size for an imperfect production system with quality assurance and rework , International Journal of Production Research, Vol.45 (14 ) ; pp.3191-3214
Chiu SW , Wang S-L , Chiu Y-SP (2007) Determining the optimal run time for EPQ model with scrap, rework, and stochastic breakdowns , European Journal of Operational Research, Vol.180 (2 ) ; pp.664- 676
Sarkar B (2012) An EOQ model with delay in payments and stock dependent demand in the presence of imperfect production , Applied Mathematics and Computation, Vol.218 (17 ) ; pp.8295-8308
Sarkar (2014) An economic production quantity model with random defective rate, rework process and backorders for a single stage production system , Journal of Manufacturing Systems, Vol.33 (3 ) ; pp.423-435
Sarkar B , Mandal P , Sarkar S (2014) An EMQ model with price and time dependent demand under the effect of reliability and inflation , Applied Mathematics and Computation, Vol.231; pp.414-421
Wee HM , Widyadana GA (2011) Economic production quantity models for deteriorating items with rework and stochastic preventive maintenance time , International Journal of Production Research, Vol.50 (11 ) ; pp.2940-2952
Sana SS ( 200) A production-inventory model in an imperfect production process , European Journal of Operational Research 201, Vol.2; pp.451-464
Sana SS ( 2010) An economic production lot size model in an imperfect production system , European Journal of Operational Research 201, Vol.201 (1 ) ; pp.158- 170
Sana SS , Chaudhuri K (2010) An EMQ model in an imperfect production process , International Journal of Systems Science, Vol.41 (6 ) ; pp.635-646
Khan M (2011) A review of the extensions of a modified EOQ model for imperfect quality items , International Journal of Production Economics, Vol.132 (1 ) ; pp.1-12
Bough Er TO (1984) Lot sizing in group technology production systems , International Journal of Production Research, Vol.22 (1 ) ; pp.85-93
Cardenas-Barron LE (2011) The derivation of EOQ/EPQ inventory models with two backorders costs using analytic geometry and algebra , Applied Mathematical Modelling, Vol.35 (5 ) ; pp.2394-2407
Barzoki MR , Jahanbazi M , Bijari M (2011) Effects of imperfect products on lot sizing with work in process inventory , Applied Mathematics and Computation, Vol.217 (21 ) ; pp.8328-8336
Ullah M , Kang CW (2014) Effect of rework, rejects and inspection on lot size with work-in-process inventory , International Journal of Production Research, Vol.52 (8 ) ; pp.2448-2460
Kim H (2012) Maximizing expected profit via multiple truck operation under imperfect truck quality , Journal of the Society of Korea Industrial and System Engineering, Vol.35; pp.45-51
Han Y , Kim H (2012) An optimal solution algorithm of the single product inventory with target in-stock ratio constant , Journal of the Society of Korea Industrial and Systems Engineering, Vol.35; pp.204-209