• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Jit kanban system review
 

Jit kanban system review

on

  • 3,023 views

 

Statistics

Views

Total Views
3,023
Views on SlideShare
3,023
Embed Views
0

Actions

Likes
0
Downloads
62
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Jit kanban system review Jit kanban system review Document Transcript

    • Int J Adv Manuf Technol (2007) 32: 393–408DOI 10.1007/s00170-005-0340-2 ORIGINA L ARTI CLEC. Sendil Kumar . R. PanneerselvamLiterature review of JIT-KANBAN systemReceived: 9 February 2005 / Accepted: 9 September 2005 / Published online: 22 March 2006# Springer-Verlag London Limited 2006Abstract In this paper, JIT (Just-In-Time)-KANBAN zero inventories, zero breakdown and so on. It ensures theliterature survey was carried out and presented. The supply of right parts in right quantity in the right place andintroductory section deals with the philosophy of JIT, and at the right time. Hence, the old system of materialthe concept involved in the push and pull system. The acquisition and, buyer and seller relationships are changedblocking mechanisms in the kanban system are also to new revolutionary concepts (Womack et al. [91],discussed elaborately. Besides these sections, the impor- Womack and Jones [92], Markey et al. [45]). Similarly,tance of measure of performance (MOP) and the applica- JIT becomes an inevitable system at plant level, whichtion of the same with respect to JIT-KANBAN are integrates the cellular manufacturing, flexible manufactur-presented. The recent trends in the JIT-KANBAN are ing, computer integrated manufacturing and Roboticsdiscussed under the heading “Special cases”. In this review, (Schonberger [63], Golhar [12]).100 state-of-art research papers have been surveyed. The Due to the technological advancement, the conventionaldirections for the future works are also presented. method of push production system linked with Material Requirement Planning (MRP) was changed to pull type JITKeywords JIT . KANBAN . Blocking Mechanisms . production system to meet out the global competition,CONWIP . Measure of performances (MOP) . Simulation where the work-in-process (WIP) can be managed and controlled more accurately than the push- production system (Mason Paul [46]).1 Introduction KANBAN system is a new philosophy, which plays a significant role in the JIT production system. Kanban isJust -In-Time (JIT) manufacturing system was developed basically a plastic card containing all the informationby Taiichi Ohno which is called Japanese “Toyota required for production/assembly of a product at each stageproduction system”. JIT manufacturing system has the and details of its path of completion. The kanban system isprimary goal of continuously reducing and ultimately a multistage production scheduling and inventory controleliminating all forms of wastes (Brown et al. [5], Ohno system. These cards are used to control production flow[54], Sugimori et al. [82]). Based on this principle, and inventory. This system facilitates high productionJapanese companies are operating with very low level of volume and high capacity utilization with reduced produc-inventory and realizing exceptionally high level of quality tion time and work-in-process.and productivity (Richard J. Tersine [62], James H. Greene The objectives of this paper are as listed below[30]). JIT emphasizes “zero concept” which means 1) Critical review of JIT literature.achievement of the goals of zero defects, zero queues, 2) Segregating the different research articles of JIT. 3) Exploring the recent trends in JIT-Kanban system and deriving directions for future research. In this paper, the articles are reviewed and an appropriateC. Sendil KumarNeyveli Lignite Corporation, classification is presented.The kanban study was madeNeyveli, India elaborately, since it acts as a basic communicator and feed- back agent to the JIT system. Push and pull system,R. Panneerselvam (*) principle of operation of kanban cards, Blocking mecha-Department of Management Studies,School of Management, Pondicherry University, nism, Toyota’s formula, and the measures of performancesPondicherry 605 014, India (MOP) are also discussed in this paper. The latest trends ine-mail: panneer_dms@yahoo.co.in JIT-Kanban system are also addressed separately under the
    • 394heading “Special cases”. Finally, the directions for future Request for items Request for itemsresearches are presented. WS 1 WS 2 STORE2 Push and pull systems Items movement Items movement Fig. 2 Pull systemPush and Pull system are two types of production systems,which operate equally in opposite sense and have their ownmerits and demerits (Monden [50], Villeda Ramiro et al. (WK), respectively. A schematic diagram of a two-card[89]). system is shown in Fig. 4.Push system It is a conventional system of production.When a job completes its process in a workstation, then it 2.1 Operation of two-card kanban systemis pushed to the next workstation where it requires furtherprocessing or storing. In this system, the job has a job card The two-card kanban pull system which works inand the job card is transferred stage by stage according to the Assembly/Manufacturing line is elaborated byits sequence. In this method, due to unpredictable changes Panneerselvam [56], Edward J. Hay [17], Kimura andin demand or production hinder-ness, the job happens to Terada [36], Hunglin Wang et al. [25] and Hemamalini etdeviate from its schedule and it causes accumulation of al. [21] and Shahabudeen et al. [76]. Basically it has plasticwork-in-process inventory. Hence, inventory planners cards, which give information about the parts and alsopessimistically fix the safety stock level on the higher things to be done. The production order kanban (POK) is aside. A schematic representation of the push system is production order, which instructs the preceding work-shown in Fig. 1. In Fig. 1, WSj is the jth workstation and station to produce the required number of units. Thethe product line consists of n workstations. withdrawal kanban (WK) gives the message to the succeeding process about the number of units it shouldPull system A pull type production system consists of a withdraw.sequence of workstations involving value addition in The simple steps adopted in kanban system are aseach workstation (WS). In the pull system, from the followscurrent workstation (j), each job is withdrawn by its 1) The container of the succeeding workstation j+1 issucceeding workstation (j+1). In other words, the job is moved to the preceding workstation j with thepulled by the successive workstation instead of being withdrawal kanban (WK) and placed it in its outputpushed by its preceding workstation. The flow of parts buffer.throughout the product line is controlled by Kanban Cards 2)(Turbo [87]). In practice, these kanban cards can be either a) Consequently it pulls the parts from output buffer of“single-card system” or “two-card system”. Each work- the workstation j and detach the production orderstation has an inbound stocking point and an outbound kanban (POK) attached to those parts and then placesstocking point. The primary advantage of the pull system the POK in the POK-post of the workstation j.is the reduced inventory and hence the associated cost of b) Work station j starts its production as per theinventory reduction. A schematic view of the pull system production order in its POK post.with two workstations and store is shown in Fig. 2. A kanban system operates only with single card iscalled production order kanban (POK) (J. Berkley [4], 3) The container along with the parts and WK movesSarathapreeyadarishini et al. [78]). If the distance between again to its succeeding workstation j+1. Then itthe consecutive workstations is very short, a single buffer delivers the parts to the input buffer of the workstationmode is made available between the workstations. This j+1 and places the WK to the WK-post of thebuffer mode acts as both outbound buffer for the current workstation j+1.workstation j and inbound buffer for the succeedingworkstation j+1, respectively. A schematic diagram of asingle-card system is shown in Fig. 3. In the two-card POK Only card movementsystem, where the distance between the two consecutive POSTwork stations are more, each work station will haveseparate inbound buffer and outbound buffer (Kimura O. etal. [36], Hemamalini et al. [21]) and the cards are called asProduction Order Kanban (POK) and Withdrawal Kanban Card + Parts movement WSj WS j+1 WS1 WS2 ooo WSj ooo WSn STORE BUFEERFig. 1 Push system Fig. 3 Schematic diagram of a single card system
    • 395 POK WK 3.1.1 Blocking due to part type POST POST (j) (j+1) This type of blocking occurs due to restriction in the number of parts (containers) that can be stored in the buffer POK WK between workstation j and the workstation j+1. The workstation j will not process the particular part p, since there is no reserved space in the buffer storage for the WS WK+ Parts WS particular part type. j j+1 Let Q(p, j, j+1) be the maximum number of units Output Buffer Input Buffer (container) of part type p that can be stored in the buffer of of storage between the workstation j and the workstation j+1. Workstation j Workstation j+1 Then the workstation j can process the p type parts, only if the actual number of units (container) of the part type p in the WK buffer storage less than Q(p, j, j+1); otherwise, the work- station j is blocked due to part type p alone. The workstationFig. 4 Schematic diagram of two card system can process any other part type provided that workstation is not blocked with respect to that part type.4) When the parts in the containers of the workstation j+1 are fully used, then the steps from 1 to 3 are repeated. 3.1.2 Blocking due to queue size3 Blocking mechanisms This type of blocking occurs due to restriction in the total number of containers of all part types in the buffer betweenEach workstation of a production/assembly line requires workstation j and the workstation j+1. The workstation jsufficient space for storing parts in its output buffers. When will not process any part type if there is no space in thethe buffer capacity of a workstation is fully occupied, no buffer storage between the workstation j and the work-further storage is possible. Because of this fact, the station j+1, irrespective of part type and container.workstation can not release the parts and hence, it can Let Q (j, j+1) be the maximum number of containersnot process components. This condition is called “Block- irrespective of the part types that can be stored in the buffering”. The blockings are categorized according to the types storage between the workstation j and the workstation j+1.as presented in Table 1. Then the workstation j can process part types, only if the actual total number of containers in the storage between the workstation j and the workstation j+1 is less than Q(j, j+1);3.1 Single card -instantaneous Otherwise, the work station j is said to be blocked due to the queue size constraint.As discussed earlier, if the workstations are situated closerto each other, the output buffer of the workstation j and theinput buffer of the workstation j+1 are one and the same. 3.1.3 Dual blocking mechanismUnder such situation, a single card instantaneous kanban isused. Berkley [4] and Sharadhapreeyadarishini et al. [77] If both the above blocking mechanisms operate simulta-have discussed the blocking mechanism of single card type neously, then it is called Dual blocking mechanism.in detail. The work station j is said to be blocked if the actual number of units (containers) of the part p in the buffer storage between the workstation j and the workstation j+1 is equal to Q(p, j, j+1) and the actual total number of containers in the buffer storage between the workstation j and the workstation j+1 is equal to Q(j, j+1).Table 1 Categories of blocking mechanisms Subsequently, when a container of the part type p is taken by workstation j+1, then the blocking is released andSingle Card -Instantaneous Two Card - Non Instantaneous the workstation j can start processing the part p. If the work1) Blocking due to part-type. 1) Blocking due to part-type. station j+1 takes the container of any part other than that of2) Blocking due to queue size. 2) Blocking due to queue size. p, then the work station j is still blocked with respect to part3) Dual blocking mechanism. 3) Dual blocking mechanism. p and it is not blocked with respect to other part types. Blocking mechanism Operative on Material Handling. 4) Blocking due to part-type. 3.2 Two card- non-instantaneous 5) Blocking due to queue size. 6) Dual blocking mechanism. If the distance between consecutive workstations is more, there will be independent input and output buffer points for
    • 396each workstation. In this system, the blocking can occur 3.3 Blocking mechanisms operative on materialdue to stagnation of parts in the output buffer of that handlingworkstation. Berkley J. [4] et al. [21] have studied this typeof blocking. Material handling operation between the workstation j and the workstation j+1 can be blocked due to part type, queue size or both. This is similar to the above types but the3.2.1 Blocking due to part type blocking is due to Material Handling (MH) between output buffer of the workstation j and the input buffer of theThis type of blocking occurs due to restriction in the workstation j+1. This was studied by Berkley J. [4] andnumber of parts (containers) that can be stored in the output Hemamalini et al. [21].buffer of the workstation j. The workstation j can notprocess the particular part p, since there is no reservedspace for the part type p in the output buffer of the 3.3.1 Blocking mechanism due to part typeworkstation j. Let Q(p, j) be the maximum number of units (containers) This type of blocking occurs due to restriction in theof the part type p that can be stored in the output buffer number of parts (containers) that can be stored in the inputstorage of workstation j. Then the workstation j can process buffer of the workstation j+1.the parts of the part type p, only if the actual number of Let M(p, j+1) denotes the maximum number of unitsunits (containers) of p in the output buffer storage of the (containers) of part type p that can be stored in the inputworkstation j is less than Q(p, j); otherwise, the workstation buffer storage of the workstation j+1.j is blocked due to the part type p alone. The workstation j Then, materials handling is permitted from the outputcan process parts of any other part type provided that the buffer of work station j to the input buffer of workstationworkstation is not blocked with respect to that part type. j+1, if the actual number of units (containers) of part p in the input buffer of work station j+1 is less than M(p, j+1).3.2.2 Blocking due to queue size 3.3.2 Blocking mechanism due to queue sizeThis type of blocking occurs due to restriction in the totalnumber of containers of all part types in the output buffer of This type of blocking occurs due to restriction in the totalthe workstation j. The workstation j will not process any of number of containers of all part types that can be stored inthe part types since there is no space in the output buffer the input buffer of the workstation j+1.storage of the workstation j, irrespective of part type and Let M(j+1) be the maximum number of containerscontainer. irrespective of part types that can be stored in the input Let Q(j) denotes the maximum number of containers buffer storage of workstation j+1. Then, the materialirrespective of part type that can be stored in the output handling is permitted from output buffer of the workstationbuffer storage of the workstation j. j to the input buffer of workstation j+1 only, if the actual Then the workstation j can process parts only if the total number of containers in the input buffer of workactual total number of containers in the output buffer of the station j+1 is less than M(j+1).workstation j is less than Q(j); otherwise, the workstation jis said to be blocked due to the queue size constraint. 3.3.3 Dual blocking mechanism3.2.3 Dual blocking mechanism If both blocking mechanisms discussed in sections 3.3.1 and 3.3.2 operate simultaneously then it is called dualIf both of the above blocking mechanisms operate simul- blocking mechanism. The material handling operation istaneously, then it is called dual blocking mechanism. said to be blocked, if the actual number of units (contain- The workstation j is said to be blocked if the actual ers) of part type p in the input buffer of the workstation j+1number of units (containers) of part type p in the output is equal to M(p, j+1) and the actual total number ofbuffer of workstation j is equal to Q(p, j) and the actual total containers in the input buffer of the workstation j+1 isnumber of containers in the output buffer of workstation j is equal to M(j+1). Subsequently, when a container of the partequal to Q(j). type p is taken from the input buffer of workstation j+1, the Subsequently, when a container of part type p is taken to blocking will be released and the material handling starts tothe input buffer of the workstation j+1, the blocking will be clear the parts from the output buffer of the work station j.released and the workstation j can start processing the part Now, the workstation j can start processing the part p. If thetype p. If the input buffer of workstation j+1 takes a workstation j+1 takes a container of parts other than that ofcontainer of parts other than the part p, then the workstation part p from the input buffer of the workstation j+1, then thej is still blocked with respect to the part type p. material handling is not possible for the part p. So the workstation j is continued to be in blocked state with
    • 397respect to the part p. However, the workstation j is not Table 2 Factors used by various researchersblocked with respect to other part types. Sl. Factors used The reference numbers No for MOP of research articles which use the MOP4 Toyota’s kanban formula 1 Average WIP [2, 6, 15, 48, 76, 79, 89, 90, 98]The formula used by Toyota Motor Company to determine 2 Demand [6, 15, 76]the number of kanbans is called Toyota formula. (Berkley 3 Fill Rate [6][4], Chan [6], Henry et al. [10], Hunglin Wang et al. [25], 4 Average kanban [74, 77, 98]Ohno et al. [53], Monden Y. [50], Philipoom et al. [60] and waiting/queue timeYavuz et al. [98]). The Toyota’s kanban formula is 5 Average Flow [2, 6, 22, 31, 51, 61, 76, 77, 98]presented below. (production lead) time 6 Average setup/process [98] D Lð1 þ αÞK time ratio C 7 Average input/output [79, 89] inventorywhere, 8 Mean cumulative [48, 74, 88, 90, 98]K is the number of kanbans, throughput rateD is the demand per unit time, 9 Mean line utilization [48, 89, 98]L is the lead-time, 10 Mean demand [98]α is the safety factor and satisfaction lead timeC is the container capacity 11 Mean staging delay of job [49, 51] From these literatures, it was noted that the lead-time 12 Mean Tardiness [2, 22, 51, 61, 79]includes waiting time, processing time, conveyance time 13 Weighted earliness of the [22, 61]and kanban collecting time. The safety stock serves as a jobbuffer against variations in both supply and demand. Henryet al. [10] has suggested some practical values for thevariables C and α. The value of C is limited to a maximum frequently used as performance measures. Some importantof 10% of demand and α is a policy variable, which is definitions for the factors, which are used in differentdecided by the management up to 10% of the demand. The MOPs by various researchers, are discussed below.variable K is the number of kanbans, which is related to the Yavuz and Satir [98] have used seven factors in theirstock. If the value of K increases, the stock of the parts also study, which are as presented below.increases. As a result, idle stock occurs. Similarly, if the 1) Mean Cumulative Throughput Rate: It is the ratio ofvalue of K decreases, the stock of parts also decreases and total satisfied demand to the total generated demand.shortage occurs. Hence, the JIT production system applies 2) Mean Total Production Lead Time: It is the amount oftrade-off between the above parameters to find the optimum time spent by a job from entering the system to untilnumber of kanbans. Many researches have been carried out completion of all operations, averaged over allto find the optimum number of kanbans using different completed job.methodologies and tools such as simulation, queuing 3) Mean Total Demand Satisfaction Lead Time: It is themodels, mathematical models, Artificial Intelligent ap- time interval between arrival of the demand andproach and so on. From the Toyota’s empirical equation, satisfaction of the demand.one can find the number of kanbans required for the system. 4) Mean Utilization of Line: It is the mean utilization of the last station in the line. 5) Mean Setup/Run Time Ratio of Line: It is the ratio5 Measure of performance (MOP) between the setup time and the run time of last station. 6) Mean Total WIP Length: It is the mean of all in-For any system, the efficiency is measured through a process-inventory levels for the products excludingfunction of related parameters/ factors. Hence these factors finished goods (FG).must obviously establish close relationship with the 7) Mean Total Waiting Time: It is the waiting time of allfocused problem. These factors individually or jointly repre- products in all processes and finished goods inventorysent a performance. Blair Berkly J. [4] has given a note on (FGI).workstation performance in kanban controlled shops interms of average inventories, quality and the ability to meet A general purpose analytical model to evaluate thethe demands. Our study reveals that various researchers performance of multistage kanban controlled productionhave used thirteen factors and they are shown in Table 2. system was developed by Di Mascolo et al. [15]. The From Table 2, it is inferred that the average work-in- performance measures used by them are percentage ofprocess (WIP), average flow time, mean cumulative demand for back-order, average waiting time of back-throughput rate and weighted earliness of the job are order and average work-in-process.
    • 398 A simulation experiment to evaluate the relative effective- Table 3 Results of the study by Chan [6]ness of various rescheduling policies in capacity-constrained, MOP Pull Hybrid HybridJIT make-to-stock production environment is examined by (Single (Single (Multi-product)Kern et al. [34]. Three performance measures analyzed by product) product)them are average finished goods inventory, total units of saleslost, and measure of schedule instability. Jing-Wen Li [31] 1) Fill rate Decrease Decrease Increasehas measured three factors for shop performance which are 2) In-process- Increase Increase Increaseaverage work-in-process (WIP) inventory, average flow time inventoryand average set up time to processing time ratio (ASOTR), 3) Manufacturing Increase Increase Decreasewhich is the ratio of total amount of time spent for setting up lead timemachines to the total amount of time spent for processingparts averaged over all machines. Uday S. Karmarker [88]used throughput rate for total work performance. In anotherstudy, the priority rule assignment was checked by the 6 Literature reviewfollowing factors by Nabil R. Adam et al. [51]. Golhar et al. [12] have classified the JIT literature as1) The lead time of a job elimination of waste, employee participation, supplier2) The flow time of the job participation and total quality control.3) The staging delay of a job A similar work was done by Berkly [4] for kanban4) Mean Tardiness production process. He has selected 24 elements in the Hemamalini et al. [22] considered the objective function kanban production system as operational design factors.to minimize the sum of weighted flow time, weighted In this section, the different topics associated with “JIT-earliness of jobs and weighted tardiness of containers. KANBAN” studied by various researchers have beenShahabudeen et al. [76] used an universal test which may grouped and presented as shown in Fig. 5. The Table 4be suited for the MOP in any JIT system, which are shows the reference numbers of the articles with respect topercentage zero demand (PZD), mean lead time (MLT) and the classifications shown in Fig. 5.mean total WIP (MTW) as explained below. Obviously, most of the researchers were focusing on the determination of number of kanbans and determining1) Percentage zero demand: It is the percentage of total corresponding solutions by using suitable models and demand immediately satisfied to the total generated tools. Some authors have developed simulations model and demand. meta-heuristics like, genetic algorithm (GA), tabu search2) Mean lead time: It is the sum of the waiting time, (TS), and simulated annealing (SA) for JIT-Kanban for processing time and moving time averaged per station. better solutions. It is also called as mean flow time. The Table 5 shows the number of articles dealt in3) Mean total WIP: It is the average number of kanbans different periods. From, Table 5, it is clear that, during last waiting for each part type at each workstation. two 5 years period (1996-2000 & 2001-2005), the number Here, PZD is a maximization measure and, MLT and of researches are more. Further, more researches have beenMTW are the minimization measures and hence the sum of done in empirical theory, flow shop, simulation, variabilitythe objective MOPs is changed as Zmax (a1PZD +a2 RMLT and its effects, CONWIP and special cases. Many+a3 RMTW), where a1, a2 and a3 are weights of the researchers have worked in JIT system with differentrespective measures and, RMLT and RMTW are modified objectives. Here, the authors have grouped some importantvalues of MLT and MTW, respectively. Chan F.T.S. [6] has done a work on how the MOP JITchanges in different production systems, while increasingthe kanban size. The measures of performance taken by SPECIAL CASEShim are as listed below1) Unsatisfied order, which is the difference between the KANBAN CONWIP actual number of unit produced and the level of SCM demand.2) Manufacturing lead-time, which is the time between POLCA the customer order and the completion of order. VARIABILITY &3) In-process-inventory is the total number of work-in- ITS EFFECTS process (WIP) inventory in units excluding finished FLOW SHOP DIFFERENT MODELS goods inventory. ASSEMBLY MATHEMATICAL4) Fill rate is the percentage of demand satisfied. LINE QUEUING MARKOVIANS SIMULATIONS The results of his study as a function of the kanban size BATCH COST MINIMIZATIONare shown in Table 3. Fig. 5 Flowchart showing the classification of literature review
    • 399Table 4 Details of classification of review articles Table 6 Objective based classification and their referencesArea of Research Reference numbers of related Articles Classification Reference numbers of articlesJIT [4, 12] A. Principles of [21, 50, 55, 59, 96]Kanban-Empirical theory [10, 33, 50, 59, 71, 93, 95] JIT-Kanban system Flow shop [7, 22, 58, 61, 77, 78] B. Operating Factors [4, 12, 80] Assembly line [16, 89, 94] C. Design of Kanban [1, 10, 13, 19, 26, 52, 53, 66, 71, 74, 75, Batch Production [35, 86] System 93] System D. Performance [7, 24, 33, 36, 48, 58, 73, 81, 89, 90, 98,Modeling Approach: [3, 36] behaviour 99] Mathematical E. Sequencing [18, 22, 57, 58, 61, 77, 78, 94] Queuing [73, 99] & Scheduling Marko-chain [14, 28, 52, 90] F. Inventory/Buffer [3, 68, 86] Simulation [1, 9, 13, 19, 26, 66, 67, 74, 75] Control Cost minimization [53, 68, 72]Variability and its effects [7, 24, 48, 89, 98] CONWIP [8, 44, 55, 69, 70, 79, 96] POLCA [69] Karmarker and Kekre [33] have concluded from their SCM [18, 29, 47, 80] studies that the reduction in container size and increase in Special Cases [6, 38, 40, 41, 43, 60, 69, 84, 85, 97, 100] number of kanbans lead to better results. Many researchers were interested in finding the optimal number of kanbans. The Toyota formula is very much useful in determining the optimal number of kanbans.objectives of the researches into six headings as shown in Co Henry et al. [10] used the Toyota formula and alsoTable 6. From Table 6, it is clear that the following investigated the safety stock allocations in an uncertainobjectives attracted more researchers. dynamic environment. A similar work was considered by Sarkar et al. [71] to find number of kanbans between two– Design of kanban system adjacent workstations. Yale T. Herer et al. [95] presented a– Performance behaviour study for kanban system, CONWIP and buffered produc-– Sequencing and scheduling tion lines. In this study, they incorporated a non-integral approach using simulation. The use of non-integral approach helps production planners to obtain discrete6.1 Empirical theory number of kanbans. Woolsey et al. [93] have developed a simple spreadsheetIn the paper by Monden Y. [50], a comprehensive optimization program to determine the correspondingpresentation of Toyota production system is given. A number of kanbans with respect to user-defined safetysuccessful kanban system will drastically reduce the stock levels and other values. It gives a close-form ofthroughput time and lead time (Philipoom et al. [59]).Table 5 Details of researches in different periodsArea of Research 1980-1985 1986-1990 1991-1995 1996-2000 2001-2005 TotalJIT 2 2Kanban- Empirical theory 1 2 3 6 Flow shop 1 4 1 6 Assembly line 1 1 1 3 Batch 1 1 2Modelling Approach: Mathematical 1 1 2 Queueing 2 2 Markovians 1 2 1 4 Simulation 1 1 5 2 9 Cost minimization 1 1 1 3Variability and its effects 1 1 3 5CONWIP 2 3 2 7 POLCA 1 1SCM 4 4Special cases 1 1 2 7 11 Total 3 10 11 23 20 67
    • 400solution to the problem. This means that an answer for any and bound algorithm, and simulated annealing algorithmproblem size may be instantaneously obtained. for finding the optimal solution and sub-optimal solution of the mixed-model sequencing problem, respectively to minimize the total conveyor stoppage time. The branch-6.1.1 Flow shop and-bound method was devoted to find the optimal solution of small-sized problems, whereas the simulatedKanban system is widely implemented in repetitive annealing method was used to cope with large-scalemanufacturing environment. For a single card operational problems to obtain a good sub-optimal solution. Future,system, Sharadhapriyadarishini et al. [77] have developed research on simulated annealing applied to this problemtwo heuristics and proved that these are more efficient. can be directed to establish a better seed generationSaradhapriyadarishini et al. [78] have proposed a recursive algorithm. However, the practitioner should spend con-equation for scheduling the single card kanban system with siderable time in fixing the parameter called temperaturedual blocking. They proposed a heuristic with twin (T) in the simulated annealing algorithm by trail and errorobjectives of minimizing the sum of total weighted time method before actually solving the problem.of containers and weighted flow time of part-types.Rajendran [61] has done a work on two card flow shopscheduling with n part-types. In this paper, mathematical 6.1.3 Batch production systemmodels for time tabling of containers for different problemshave been formulated. Then, a heuristic was developed to In a batch production system, the switching over from oneminimize the sum of weighted flow time, weighted earliness, product to other product depends on many factors such asand weighted tardiness of containers. Hemamalini et al. [22] stock reaching to the threshold level, different priorityhave done similar work. In this work, the heuristic developed schemes, economical setups, etc. Tafur Altiok et al. [86]is simulated annealing algorithm. This is compared with have dealt this issue differently for the pull typerandom search method. In these papers, the comparisons are manufacturing system with multi product types. In thisdone only based on mean relative percentage increase. paper, they developed an iterative procedure to approxi-Instead of this approach, comparisons based on complete mately compute the average inventory level of eachANOVA experiments would provide reliable inference. product as finished goods using different priority schemes. Peter Brucker et al. [58] have carried out research on In this paper, the demand arrival process is assumed to beflow shop problem with a buffer of limited capacity a poisson distribution and processing times and the set-upbetween two adjacent machines. After finishing the times are arbitrarily distributed. But, in practice, theprocessing of a job on a machine, either the job is to be processing times may follow other distributions, viz.,processed on the following machine or it is to be stored in normal, uniform, exponential, etc. which are not experi-the buffer between these machines. If the buffer is mented in this paper. Khan et al. [35] addressed thecompletely occupied, the job has to wait on its current problem of manufacturing system that procures rawmachine but blocks this machine for other jobs. In this materials from vendors in lot and convert them intopaper, they determined a feasible schedule to minimize the finished products. They estimated production batch sizesmakespan using tabu search. The results of the problem for JIT delivery system and designed a JIT raw materialusing tabu search were compared with that of benchmark supply system. A simple algorithm was developed toinstances. The comparisons are done only based on relative compute the batch sizes for both manufacturing and rawimprovements. Instead of this approach, comparisons material purchasing policies.based on complete ANOVA experiments would providereliable inference. 6.2 Modeling approach6.1.2 Assembly line Modelling approach aims to obtain the optimal solution. This subsection reviews different modeling approaches.Assembly lines are similar to the flow shops in whichassembly of parts are carried out in a line sequence. In amulti product assembly line, the sequencing of the jobs is a 6.2.1 Mathematical modelchallenging task. Drexl et al. [16] considered an assemblyline sequencing mixed model problem. It is a combinatorial Kimera and Terada [36] have developed a mathematicalproblem. They formulated this combinational problem as model in the area of kanban system. They have given ainteger programming model. This model can be used only basic balance equation for multi stage systems, whichfor small size problems due to the limitations of operations shows how the fluctuation of final demand influences theresearch software with respect to handling the number of fluctuation of production and inventory volumes. Bitranvariables and constraints, which are present in the integer- and Chang [3] have designed an optimization model for theprogramming model. Xiaobo et al. [94] have considered kanban system. The model is intended for a deterministicsimilar work on mixed model assembly line sequencing multi-stage capacitated assembly-type production setting.problem with conveyor stoppages. They proposed branch In this paper, a non-linear model developed by them is
    • 401converted into a linear model with deterministic demand. approach involves 3 steps methodology, viz., 1) dataThis deterministic model is designed to find the choice of collection, 2) formation of decision tree, and 3) interpre-the number of kanbans to be used at each stage of a given tation of decision tree. This method helps to set kanbanproblem and to control the level of inventory. But this levels under high demand variability. The results show thatanalysis does not include uncertainties directly. Hence, the rule induction using CART is a viable solution to theutility of this model is very much limited. knowledge acquisition bottleneck. Hence, an extended work on knowledge acquisition for this domain will be a significant contribution to literature.6.2.2 Queuing modelSeki et al. [73] have designed a single-stage kanban system 6.2.4 Simulation based studieswith poisson demand arrivals. The system is formulated asa queuing system under piecewise constant load, and a There are many simulation softwares available in thenumerical method by transient solutions of the queue is market, such as GPSS, Q-GERT, SLAM-II, SIMAN,applied. This method, which shows the transient behavior SIMSCRIPT, EXTEND, ARENA, and SIMULINK. Sim-of the kanban system, gives a better result. Yoichi Seki et ulation uses the attributes/parameters of a problem to arriveal. [99] did similar work on the single stage kanban system the results. As for as designing of kanban system, a basicwith poisson demand and erlang production times. The simulation study was done by Davis et al. [13] and Gabrielobjective of this work is to determine the number of et al. [19] to determine the number of kanbans. In anotherkanbans, when a change of load to the system is planned. work by Rudi De Smet et al. [67], a simulation model wasThey mainly proposed a numerical method by transient developed to study the feasibility of plans to produce somesolutions of the queueing system which was developed subparts of the product in a kanban-controlled manner tounder piecewise constant load. This method also shows determine the operational parameters such as number ofthat the transient behavior of the kanban system operates kanbans and container size. This feasibility study wasbetter with other parameters. In this paper, the load carried out for two situations, namely (1) all subpart typesdistribution is assumed to be piecewise linear. Instead, it are produced in a kanban controlled manner and (2) onlycan be assumed as a continuous distribution and the the production of fast-movers on two (out of three)corresponding results using simulation can be compared machines is kanban controlled. The result assures that thewith `the results of this paper. kanban control is the best method for fast moving parts. In a kanban control system, the main decision parameters are the number of kanbans and lot size. Alabas6.2.3 Markovians model et al. [1] developed three-meta heuristics viz., genetic algorithm (GA), simulated annealing (SA) and tabu searchVito Albino et al. [90] studied a model of kanban controlled (TS) coupled with a simulation model to find the optimummanufacturing system based on Markovian assumption. number of kanbans with the minimum cost. In addition, aAn approximate approach was developed to solve the neural network metamodel was developed and comparedmodel, which permits reliable evaluation of performance in with the heuristic procedures in terms of solution accuracy.terms of throughput time and work-in-process (WIP). They found that the tabu search requires less computationalFurther, they validated the results using discrete-event efforts when compared to the other two meta-heuristics andsimulation applied to their problem. It was observed that the neural network meta-model. In a similar work bythe results of the approximation approach did not deviate Hurrion R.D. [26], simulation and neural network meta-much from that of the simulation approach. The errors were model have been used for designing the kanban system. Inalways within 5% even for moderate size problems with 20 this paper, an approximate solution is found using neuralstages. The comparisons made in this paper were based on network meta-mdoel and then it is used as the starting pointabsolute value of percentage relative errors. Instead of this in simulation to find the optimum number of kanbans of aapproach, they should have done comparisons through a manufacturing system. Actually, the word “optimum”carefully designed ANOVA experiments. Nori and Sarkar should have been avoided in his paper, because neither[52] have modeled the kanban system using Markov-chain the proposed meta-model nor the simulation approach willto determine the optimum number of kanbans between give optimal number of kanbans. The optimum number ofadjacent workstations. kanbans may be called as the minimum number of kanbans. Deleersnyder et al. [14] have modeled a blocking In this context, an attempt has been made bysituation in the queues of the kanban system using discrete Shahabudeen et al. [75] to set the number of kanbans astime Markovian chain to study the effect of number of well as lot size at each station using simulated annealingkanbans, machine reliability, processing time and demand algorithm. A simulation model with a single-card systemvariability. Markham et al. [28] formed a procedure based has been designed and used in the analysis. A bi-criterionrule induction approach for determining the number of objective function comprising of mean throughput rate andkanbans and other factors in JIT. They applied classifica- aggregate average kanban queue, has been used fortion and regression tree (CART) technique to generate the evaluation. In another work of them (Shahabudeen andproduction rule, based on decision trees. This system Krishnaiah [74]), they have set the number of production
    • 402kanbans and withdrawl kanbans at each workstation, and optimal JIT buffer level is determined from a cost analysislot size using genetic algorithm (GA). The solution of the using trade-off between the holding cost per unit of timegenetic algorithm is found to be better than the random and the shortage cost per unit time such that their sum issearch procedure. They concluded that the genetic algo- minimized (Salmark et al. [68]).rithm gives better solution for the assumed kanban system. A paper by Royce O. Bowden et al. [66] describes theuse of evolutionary programming (EP) integrated with a 6.3 Variability and its effectssimulation model of manufacturing system to determinethe minimum number of kanbans and corresponding Mehmet Savsar et al. [48] studied a simulation model toproduction trigger values required to meet the demand. In investigate the effect of different operational conditions,this paper, the inference is drawn for each measure, based including kanban withdrawal policies on three perfor-on single replication under each solution-technique. The mance measures of JIT, viz., average throughput rate,authors could have designed a single factor ANOVA average station utilization and average work-in-process.experiment for each measure in which “Solution Tech- Unlike other simulation studies that use exponential ornique” as the factor, with desirable number of replications truncated normal distribution, this model uses Erlang andto obtain reliable inference of their simulation study. Gama distribution. It is observed that the throughput rate asChristos G. Panayioton et al.[9] have developed a simu- well as the average station utilization is significantlylation based algorithm for determining the minimum affected by the variability in processing time and demandnumber of kanbans in a serial production system in order intervals. They proposed two types of kanban withdrawalto maximize the throughput rate and minimize work-in- cycles, namely fixed withdrawal policy and variableprocess inventory. The finite perturbation analysis (FPA) withdrawal policy. Under the fixed withdrawal policy, thetechnique was used in the simulation and to get sensitivity time interval between consecutive visits of a part-carrier toresults. They have considered single product in the a workstation for kanban removal is fixed, but the orderproduction line. But, in most of the cases, production quantity (number of kanbans carried) is variable whereaslines will be manufacturing multi-products. The assump- under the variable withdrawal policy, the time intervaltions of arbitrary arrival and service process distributions between consecutive visits of a part-carrier to a workstationlimit the scope of application of this paper in practice. for kanban removal is variable, but the order quantity is fixed. As an extension of this work, the effects of different combinations of the two kanban withdrawal policies and6.2.5 Cost minimization model number of kanbans between workstations, on the perfor- mance measures can be compared.Ohno et al. [53] proposed an algorithm to determine the Huang et al. [24] have found that overtime required willoptimal number of kanbans for each of the two kinds of be increased when the variation in processing time iskanban (production ordering and supplier kanbans) under increased. Also, they emphasized that a kanban systemstochastic demand. An algorithm was devised for deter- would not be effective with high variable processing or setmining the optimal number of kanbans that minimizes the up time. Villeda et al. [89] performed a simulation study forexpected average cost per period. Since, no safety stock is a final assembly consisting of “3 sub-assembly lines and 4assumed in this paper, this can be regarded as a procedure stages” repetitive production systems with kanbans. Theyfor determining the safety stock also. Sarkar et al.[72] concluded that improved productivity obtained throughstudied a multi stage kanban system for short life-cycle unbalancing the processing time at all workstationsproduct in the market. In this research, the problem is to increases directly with the variability in the final assembly.find optimally the number of orders for raw-materials, Chaturvedi and Golhar [7] simulated a kanban based flowkanbans circulated between workstations, finished goods production line for a product in nine sequentially arrangedshipments to the buyers, and the batch size for each workstations. They observed that the system performanceshipment (lot) with minimum total cost of the inventory. A was worst for exponential processing time distribution andcost function was developed based on the costs incurred for variability affected station utilization, throughput time andthe raw materials, the work-in-process and the finished WIP inventory. Yavuz and Satir [98] have studied thegoods. The optimal number of raw material orders that simulation of multi-item, multi-stage flow line operatingminimizes the total cost is obtained first, which is then used under the JIT philosophy with a two-card kanban tech-to find the minimum number of kanbans, finished goods nique. The flow line produces four products through fiveshipments, and the batch sizes of shipments. This paper stations. This study uses partial factorial design fordiscusses a stage-wise optimization. Instead, a fully experimentation. Seven experimental clusters are designed,integrated approach may be followed. Further, this paper each composed of at most three factors. The F ratios andconsiders single product, with constant production rate at the degrees of freedom of the model are obtained fromeach workstation in a serial production line. So, the work multi-variate analysis of variance (MANOVA). They foundmay be extended for multi-product with varying production that decrease in lot size reduces mean length and waitingrate at each workstation in an assembly-type production. times in work-in-process points at all kanbans levels. An During preventive maintenance, a JIT buffer is needed increase in the uncertainty of demand arrival rates andso that the normal operation will not be interrupted. The demand sizes increases the probability of sudden over-
    • 403loading. An increase in the coefficient of variation in as-needed basis only, and production begins only whenprocessing times brings about higher line utilization and a requested. It is supposed to match customer demand, thatdecrease in throughput rate. The scheduling rules tested is, producing only enough to replenish what the customerin this paper are found to yield no significant differences in has used or sold.the utilization of line and on the behaviours of work-in- F. Elizabeth Vergara et al.[18] have dealt the co-process. Feeder lines may be introduced into the pull ordination between different parts of simple supply chain.system configuration, where lines feed the final assembly Materials should be moved from one supplier to otherline. Further, alternate operating routes for the products supplier as per the JIT. For this, an evolutionary algorithmalong the line may be introduced. was used which identifies the optimal or near optimal, synchronized delivery cycle time and suppliers’ compo- nent sequences for a multi-supplier, multi-component6.4 CONWIP simple supply chain. The evolutionary algorithm also calculates a synchronized delivery cycle time for the entireCONWIP is a kanban system working with constant work- supply chain, the cumulative cost throughout the supplyin-process. CONWIP is a generalized form of kanban. chain, and the cost to each supplier. The results of thisLike, kanban system, it relies on signals, which could be algorithm were compared with enumeration method andelectronic and it is equivalent to kanban cards. In a found that the evolutionary algorithm gives better solutionCONWIP system, the cards traverse a circuit that includes in quick manner. This algorithm uses only two-pointthe entire production line. A card is attached to a standard crossover genetic operators. A third genetic operator maycontainer of parts at the beginning of the line. When the be introduced to further improve the performance of thecontainer is used at the end of the line, the card is removed evolutionary algorithm. The evolutionary algorithm mayand sent back to the beginning of the line where it waits in a be modified to handle complex supply chain problem.card queue to eventually be attached to another container of Stefan Minner [80] did a comprehensive review ofparts. multiple-supplier inventory models in supply chain Oscar Rubiane et al. [55] have reviewed the literatures management. SCM discusses strategic aspects of supplierand presented the benefits and comparison of the CONWIP competition, operation flexibility, global sourcing andsystems. Most of the articles reveal that the CONWIP inventory models. Further it was extended to logisticssystem works more efficiently than the conventional and multi echelon system. The emerging importance of E-kanban systems. Yang and Kum Khiong [96] compared 3 business, especially E-procurement possibilities with thedifferent systems viz., Single Kanban, Dual Kanban and use of Internet technologies reduces transaction costs forConwip. The results show that CONWIP consistently supplier search and order placement with several suppliersproduces the shortest mean customer waiting time and and therefore multiple-supplier models are more attractivelowest total work-in-process. Spearman et al. [79] have when compared to single sourcing alternative. This type ofstressed that the flexibility of CONWIP system allows it to market with spot offers, continuously changing suppliersbe used by any product-line where the utility of kanban and high uncertainty with respect to lead-time andsystem is limited. Hence, the superiority of CONWIP pull reliability of supplies, makes multiple-supplier replenish-system is an alternative to kanban system. They present ment and inventory strategies outperform single sourcingtheoretical arguments and simulation study of CONWIP. policies. Matheo et al. [47] have carried out a case study on Christelle Duri et al. [8] have analyzed CONWIP inventory management in a multi-echelon spare partssystem, which consists of three stations in series. When a supply chain. This paper clearly shows the close relation-finished part is consumed by a demand, a raw part is ship between supply chain structure and demand patterns.released immediately and gets processed at each station The problems of managing supply chain with varioussequentially. The processing at each station does not numbers of echelons, multi model, extremely variablealways meet the requirement of quality. Hence, at the end demand and lack of visibility over the distribution channelof processing in a station, the part is checked for quality are discussed. They provided an algorithmic solutionand if it is not as per the standard, then it is sent back to the through the comprehension of the sources of demandsame station for reprocessing. They proposed an analytical variability and through a probabilistic forecast and inven-method to evaluate the performance of this kind of system. tory management. Isreal David et al. [29] have enumeratedIn this paper, only three stations in series are considered. the vendor-buyer inventory production models. They argueAs an extension, a CONWIP system with generalized, that there should be a certain degree of independencen stations in series may be analyzed. between successive links of the supply chain, to allow flexibility in production management in individual links. They identified the degree of independence and level of6.5 SCM flexibility in terms of lot sizing and delivery scheduling in a single-vendor-single-buyer system. In these lines, appro-There are number of articles in SCM (Supply Chain priate two-sided vendor-buyer inventory production mod-Management). In this present survey, a few JIT-SCM els are formulated and analyzed.related articles are reviewed. In pull production manage- In all the papers, simulation as well as meta-heuristicsment systems such as JIT, deliveries must be made on an can be used as powerful tools to derive results under
    • 404probabilistic conditions. Future research can be focused in under various stable-demand conditions. The performancethese lines. of this system shows superior result. Philipoom et al. [60] have done a kanban design with flexible Toyota’s system. This system can dynamically6.6 Special cases adjust the number of kanbans at each workstation in an unstable production environment according to need/Sarah M. Rayan et al. [69] have defined POLCA system. demand and lead time to reduce the cost. The work ofPOLCA stands for Paired Cell Overlapping Loop of Cards Tardif et al. [85] introduces a new adaptive kanban-typewith Authorization. This system assumes that the factory pull control mechanism which determines the timings tohas been partitioned into non-overlapping manufacturing release or reorder raw parts based on customer demandscells. POLCA maintains constant WIP (like CONWIP) and inventory back orders. In the adaptive pull system, thebetween every pair of cells that experiences inter cell part number of kanbans in the system is dynamically readjustedmovement. Part release to a cell requires an appropriate based on current inventory level and backorder level.kanban cards as well as an authorization from the factory Unlike a conventional system, this system absorbs extraloading system. CONWIP and POLCA achieve a better kanbans according to the variability in demand. It wastrade-off between total WIP and total throughput time than found from the results of a simulation study of a single-that of other systems. Their application of single chain stage, single product kanban system that these systems areanalysis for multiple chain operation raises an open beneficial in production line under variable demandquestion whether a single WIP level should be maintained conditions. It shows that this adaptive system under suchfor all products or individual levels for each product. conditions outperforms the traditional kanban pull controlFurther, most of the studies use simulation. Hence, future mechanism. This adaptive approach may be extended forresearch should be directed to develop improved search multi-stage, multi-product kanban system.procedures for finding WIP levels in kanban systems. Kutc So [40] presents the buffer allocation problem with Krieg et al. [38] considered a kanban controlled the objective of minimizing the average work-in-processproduction system with 3 or more different products subject to minimum required throughput rate and constraintprocessed on a single manufacturing facility as a decom- on the total buffer space availability. Both the balanced andposed system. The customers for a product arrive as per unbalanced lines were considered in this work. On the basispoisson distribution. The service time and set-up changes of empirical results, he developed a good heuristic forare product specific and follow exponential distribution. If selecting the optimal buffer allocations. The mathematicalthe customer’s demand cannot be met from stock, the model discussed in this paper is based on the twocustomer leaves and satisfies his demand elsewhere (lost assumptions that there are always materials available forsales). The production run continues until the target processing at the beginning of the line and that the lastinventory level given by the kanbans for the product has station can never be blocked. But one can assume that thebeen reached. Then the manufacturing facility is set-up for first station can start processing when there are jobsproducing the next product. The basic principle of the waiting which arrived as per poisson arrival patterndecomposed method is to decompose the original multi- (make-to-order environment). In contrast, one can considerproduct system into a set of single-product subsystems (one the situation where there is finite buffer after the lastfor each product). Each subsystem is modeled approxi- station where the finished items are consumed by demandsmately by a continuous-time Markov chain. In this buffer with poisson pattern (make-to-stock environment). Un-allocation problem, the objective is to minimize the fortunately, in either case, the resulting Markov chain hasaverage work-in-process subject to a minimum required infinite number of states. So, one can develop simulationthroughput and a constraint on the total buffer space. The model as the last resort for studying the problem underresults of the decomposition method are compared with these two environments.that of exact method and simulation. The key performance Lai et al. [41] have proposed a system dynamic (SD)measures are with small relative errors for the decompo- methodology for studying the new generation of JIT insition method. As an extension to this work, a decompo- electronic commerce environment. It is a framework forsition algorithm can be developed for multi-product thinking about how the operating policies of a companykanban systems with state dependent setups. Takashashi and its customers, competitors and suppliers interact toet al. [84] have proposed a decentralized reactive kanban shape the company’s performance over time. The systemsystem for multi-stage production and transportation dynamic is a study which deals with the information feed-system with unstable changes in product demand. In the back and its evolution into future decision making. Itproposed system, the time series data of the demand from provides a new analysis of logistics policies of thethe succeeding stage are monitored at each stage company. Future work can be on extending the variablesindividually and unstable changes in the demand are and elements and to conduct experiments to investigate thedetected by utilizing control charts. In order to develop a stability of the system under various conditions such as thecontrol rule of the buffer size, the multi-stage production sudden increase in demand and random demand, experi-and transportation system is decomposed into single-stage mentation on the system behaviour of different types ofprocessing systems and the performance of the decom- customer and modes of manufacturing.posed system is investigated by simulation experiments
    • 405 The papers by Yannick Frein et al. [97] and Yves Dallery The relationship between implementation of TQM, TPMet al. [100] introduce a new mechanism for the coordination and JIT will lead to improvement in the manufacturingof multistage manufacturing system called extended kanban performance (Kribty et al. [37]). Further Huang [23]control system (EKCS). It depends on two parameters per discusses the importance of considering the integration ofstage viz., the number of kanbans and base stocks of finished TPM, JIT, Quality control and FA (Factory Automization).products. This EKCS is evolved from combining classical Imai [27] believes that TQM and TPM are the two pillarskanban and base stock control system. The advantages of the supporting the JIT production system. Kakuro Amasakaextended kanban control system were compared with the [32] proposes a new JIT management system, which helpsgeneralized kanban control system (GKCS). It was found to transfer the management technology into managementthat the capacity of EKCS depends only on the number of strategy.kanbans but not on the base stock of finished parts. Fullerton et al. [65] have conducted a study in 253 firms A work done by Chan [6] describes the practical in USA to evaluate empirically whether the degree withapproach to determine the optimal kanban size using which a firm implements the JIT practices affects the firmssimulation. The research was done basically for single financial performance. From their study, JIT manufacturingproduct and multi product manufacturing environments in system will reap sustainable rewards as measured bytwo types of JIT production systems, the pull-type and improved financial performance. Also, they studied thehybrid-type. Their measures of performance obtained benefits of JIT implementation in 95 firms in USA. Theythrough simulation models are compared. For a single have concluded that JIT implementation improves theproduct, when there is an increase in kanban size, the fill performance of the system, because of resultant qualityrate decreases whereas with both in-process-inventory and benefits, time based benefits, employees flexibility,the manufacturing lead time increase. For multi-products accounting simplification, firms profitability and reducedmanufacture, when there is an increase in kanban size, inventory level.there is an increase in the fill rate with a decrease in themanufacturing lead-time. Leyuan Shi and Shulimen [43]have presented a hybrid algorithm for buffer allocation 8 Conclusionproblem called the hybrid nested partition method (NP) andtabu search method (TS). The nested partitioned method is The growing global competition forces many companies toglobally convergent and can utilize many of the existing reduce the costs of their inputs so that the companies canheuristic methods to speed up its convergence. In this have greater profit margin. There are considerablepaper, tabu-search is incorporated in the nested partitioned advancements in technology and solution procedures inframework and it was found that such incorporation results reality, to achieve the goal of minimizing the costs ofin superior solutions. The new algorithm is efficient for inputs. JIT-KANBAN is an important system, which isbuffer allocation problems in larger production lines. The used in production lines of many industries to minimizenested partitioned method can be enhanced by incorporat- work-in-process and throughput time, and maximize lineing any one or a combination of the many other heuristics efficiency. In this paper, the authors have made an attemptviz., elaborate partitioning, sampling, backtracking to review the state-of-art of the research articles in the areascheme, simulation, etc. Then, they can be applied to “JIT-KANBAN system”. After a brief introduction to pushcombinatorial problems of this type. and pull systems, different types of kanban and their operating principles, blocking mechanisms, the authors have classified the research articles under JIT-KANBAN7 JIT integration, implementation and benefits system into five major headings, viz., empirical theory, modeling approach, variability and its effect, CONWIP andJust-in-time is a manufacturing philosophy by which an JIT-SCM. Also, the authors have provided a section fororganization seeks continuous improvements. For ensuring special cases under JIT-KANBAN. This paper would helpcontinuous improvements, it is necessary for any organi- the researchers to update themselves about the currentzation to implement and integrate the JIT and JIT related directions and different issues under JIT-KANBANareas. If it is practiced in its true sense, the manufacturing system, which would further guide them for their futureperformance and the financial performance of the system researches.will definitely improve. The directions for future researches are presented below. Swanson et al. [83] have reiterated that proper planning The flow shop as well as mixed model assembly lineis essential for implementation of a JIT manufacturing problems come under combinatorial category. Hence,system and a commitment from top management is a pre- meta-heuristics viz., simulated annealing, genetic algo-requisite. Cost benefit analysis is to be studied initially with rithm and tabu search may be used to find solution tothe knowledge of key items such as the cost of conversion determine the minimum number of kanbans and otherto a JIT system and time period of conversion. Cook et al. measures. In simulated annealing algorithm, researchers[11], in their case study for applying JIT in the continuous can aim to device a better seed generation algorithm whichprocess industry, show improvements in demand forecast will ensures better starting solution. In most of the papers,and decrease in lead-time variability. comparisons are done only based on relative improve- ments. Instead of this approach, comparisons based on
    • 406complete ANOVA experiments would provide reliable or a combination of the many other heuristics viz.,inferences. elaborate partitioning, sampling, backtracking scheme, Under batch processing system with multiple product simulation, etc. Then, they can be applied to combinatorialtypes, research may be directed to study the effect of problems of this type.different combinations of probability distributions for Ants colony optimization algorithm is a recent inclusionarrival process and processing times on the average to the existing meta-heuristics viz., simulated annealinginventory level of each product as finished goods. algorithm, genetic algorithm and tabu search. So, a As an extension to the work of Markham et al. [28] on a researcher can study the solution accuracy as well asprocedure based rule induction approach for determining required computational time of this algorithm for his/herthe number of kanbans and other factors in JIT, develop- JIT problem of interest, which falls under combinatorialment of knowledge acquisition for this domain will be a category and compare its results with the results of thesignificant contribution to literature. other three heuristics (meta-heuristics). Sarkar et al. [72] did stage-wise optimization for a multistage kanban system for short life-cycle product in the Acknowledgement The authors thank the unanimous referees formarket. This work may be extended for multi-product with their constructive criticisms, which helped them to improve the content and presentation of this review paper.varying production rate at each workstation in anassembly-type production. As an extension of the work of Mehmet Savsar et al. References[48], a study can be carried out to compare the effects ofdifferent combinations of the two kanban withdrawal 1. Alabas C, Altiparmak F, Dengiz B (2002) A comparison of thepolicies and number of kanbans between workstations, on performance of artificial intelligence techniques for optimizingthe performance measures. In the work of Yavuz and Satir the number of kanbans. J Oper Res Soc 53(8):907[98] on multi-item, multi-stage flow line, simulation model 2. Petroni A, Rizzi A (2002) A Fuzzy logic based methodology to rank shop floor dispatching rules. Int J Prod Econ 76:99–108may be developed with feeder lines introduced into the pull 3. Bitran GR, Chang (1987) A mathematical programmingsystem configuration, where lines feed the final assembly approach to a deterministic Kanban system. Manage Sciline. Further, alternate operating routes for the products 33:427–441along the line may be introduced. 4. Blair, Berkley J (1992) A review of the kanban production control research literature. Prod Oper Manag 1:393–411 As an extension to the work of Chaturvedi and Golhar 5. Brown KA, Mitchell TR (1991) A comparison of just in time[7], a CONWIP system with generalized, n stations in and batch manufacturing the role of performance obstacles.series may be analyzed. Acad Manage J 34(4):906–917 This algorithm developed by Elizabeth Vergara et al. 6. Chan FTS (2001) Effect of kanban size on just in time[18] uses only two-point crossover genetic operators. A manufacturing system. Journal of Material Processing Tech- nology 116:146–160third genetic operator may be introduced to further improve 7. Chaturvedi M, Golher DY (1992) Simulation modeling andthe performance of the evolutionary algorithm. The analysis of JIT production system. Prod Plan Control 3(1):81–92evolutionary algorithm may be modified to handle com- 8. Christelle D, Yannick F, Lee H-S (2000) Performanceplex supply chain problem. In JIT-SCM related research evaluation and design of CONWIP system with inspection. Int J Prod Econ 64:219–229works, effort should be directed to develop simulation as 9. Panayiotou C G, Cassandras C G (1999) Optimization ofwell as meta-heuristics to derive results under probabilistic kanban - based manufacturing systems. Automatica 35c:conditions. 1521–1533 In the work of Sarah M. Rayan et al. [69], the application 10. Co Henry C, Sharafali, Moosa (1997) Overlapping factor in Toyota’s formula for computing the number of kanbans. IIEof single chain analysis for multiple chain operation raises Trans 29(5):409–415an open question whether a single WIP level should be 11. Cook R L, Robert A (1996) Applying JIT principles tomaintained for all products or individual levels for each continuous process manufacturing supply chains. Prod Inventproduct. Further, most of the studies use simulation. Hence, Manage J 37(1):12–17future research shall be directed to develop improved 12. Golhar Damodar Y, Stamm C L (1991) The just in time philosophy: a literature review. International J Prod Res 29search procedures for finding WIP levels in kanban (4):657–676systems. As an extension to the work of Krieg et al. [38], 13. Davis WJ, Stubitz SJ (1987) Configuring a Kanban systema decomposition algorithm can be developed for multi- using discrete optimization of multiple stochastic responses. Intproduct kanban systems with state dependent setups. The J Prod Res 25:721–740 14. Deleersnyder JL, Hodgson TJ, Muller Malek H, O’Grady PJadaptive approach suggested by Tardif et al. [85] may be (1989) Kanban controlled pull system an analytic approach.extended for multi-stage, multi-product kanban system. Manage Sci 35:1079–1091The work of Lai et al. [41] can be extended by including 15. Di Mascolo et al (1996) An analytical method for performancemore variables and elements and conducting experiments evaluation of kanban controlled production systems. Oper Res 44(1):50–64to investigate the stability of the system under various 16. Drexl et al (2001) Sequencing JIT mixed-model assembly linesconditions such as the sudden increase in demand and under station-load and part-usage constraints. Manage Sci 47random demand, experimenting on the system behaviour of (3):480different types of customer and modes of manufacturing. 17. Hay E J (1988) The just in time break through. John Wiley &The nested partitioned method provided by Leyuan Shi and sons, New YorkShuli Men [43] can be enhanced by incorporating any one
    • 40718. Vergara EF, Moutaz K, Zhigniew M (2002) An evolutionary 43. Leyuan S, Shulimen (2003) Optimal buffer allocation in algorithm for optimizing material flow in supply chain. Comput production lines. IIE Trans 35:1–10 Ind Eng 43:407–421 44. Lambrecht M, Segaert A (1990) Buffer stock allocation in serial19. Gabrial T, Bicheno J, Galletly JE (1991) JIT manufacturing and Assembly type of production line. Int J Oper Prod Manage simulation. Industrial Management & Data system 91:4 10:220. Gravel M, Prince W (1988) Using the Kanban in a job shop 45. Markey M (1996) Examining a kanban material acquisition environment. Int J Prod Res 26(6):1105–1118 system. Ind Manage 38(3):22–2621. Hemamalini B, Rajendran C (2000) Determination of the 46. Mason P A (1999) MRP II and kanban formula. Logist Focus number of containers, production kanbans and withdrawal 7:19–23 kanbans; and scheduling in kanban flowshop- Part I. Int J Prod 47. Kalchschmidt M, Giuliozotheri, Verganti R (2003) Inventory Res 38(11):2529–2548 management in a multi echelon spare parts supply chain. Int J22. Hemamalini B, Rajendran C (2000) Determination of the Prod Econ 81 & 82:397–413 number of containers, production kanbans and withdrawal 48. Mehmet S, Adbullah-Al-Jawini (1995) Simulation analysis of kanbans; and scheduling in kanban flowshop- Part II. Int J Prod Just in Time production system. Int J Prod Econ 42:67–78 Res 38(11):2549–2572 49. Mohanasundaram KM, Natarajan K, Viswanathkumar,23. Huang P (1991) World class manufacturing in 1990s integrat- Radhakrishnan P, Rajendran C (2002) Scheduling rules for ing JIT,TQC,FA, and TPM with worker participation, modern dynamic shop that manufacture multi-level jobs. Comput Ind production concept theory and application. Springer, Berlin Eng 44:119–131 Heidelberg New York 491–507 50. Monden Y (1983) Toyota production system. Ind Eng Manage24. Huang PY, Rees LP, Taylor BW (1983) A simulation analysis press, - Atlanta of Japanese Just In Time technique with Kanban for multi line 51. Adam N R, Will J, Bertrand M, Surkis J (1987) Priority multi stage production system. Decis Sci 14:326–344 assignment procedure in multi level assembly jobshops. IIE25. Wang H, Wang H-P(Ben) (1999) Optimum number of kanban Trans 19(3):317–328 between two adjacent workstations in JIT system. Int J Prod 52. Nori VS, Sarker BR (1998) Optimum number of kanbans Econ 22:179–188 between two adjacent stations. Prod Plan Control 9:60–6526. Hurrion RD (1997) An example of simulation optimization 53. Ohno K, Nakashima K, Kojima M (1995) Optimal number of using a neural network metamodel: finding the optimum two kinds of Kanbans in a JIT production system. Int J Prod number of kanbans in a manufacturing system. J Oper Res Soc Res 33:1387–1401 48(11):1105–1112 54. Ohno T (1988) Toyota production system beyond large scale27. Imai M (1988) Will America corporate theme song be Just In production. Productivity Press, Cambridge Time ? J Qual Partic 21(2):26–28 55. Ovalle O R, Marquez A C (2003) Exploring the utilization of28. Markham Ina S et al (1998) A rule induction approach for CONWIP system for supply chain management. A comparison determining the number of kanbans in a Just in Time with fully integrated supply chain. Int J Prod Econ 83:195–215 production system. Comput Ind Eng 34(4):717–727 56. Panneerselvam R (1999) Production and operation manage-29. Israd David, Mosh Eben C (2003) How for should JIT vendor ment. Prentice Hall of India, New Delhi buyer relationship go? Int J Prod Econ 81:361–368 57. Patrick R, Mc Mullen, Frazier G V (2000) A simulated30. Greene James H (1987) Production and inventory control hand annealing approach to mixed model sequencing with multiple book - American production & Inventory control society, Mc objectives on Just in Time line. IIE Trans 32:679–686 Graw Hill Book Company 58. Brucker P, Heitmann S, Huricnk J (2003) Flow-Shop problems31. Jing-wen-Li (2003) Improving the performance of Job shop with intermediate buffers. OR-Spectrum 25:549–574 manufacturing with demand-pull production control by redu- 59. Philipoom et al (1987) An investigation of the factor influencing cing setup/processing time variability. Int J Prod Econ 84:255– the number of kanbans required in implementation of the JIT 270 technique with Kanbans. Int J Prod Res 25(3):457–47232. Kakuro A (2002) New JIT A new management technology 60. Philipoom PR, Rees LP, Taylor BW, Huang (1987) Dynami- principle at Toyota. Int J Prod Econ 80:135–144 cally adjusting the number of kanban system in JIT production33. Karmarker and Kekre (1989) Batching policy in kanban system using estimated values of lead time. IIE Trans 19 system. J Manuf Syst 8:317–328 (2):199–20734. Kern et al (1996) Master production rescheduling policy in 61. Rajendran C (1999) Formations and heuristics for scheduling in capacity-constrained just- in-time make-to-stock environments. kanban flow shop to minimize the sum of weighted flowtime, Decis Sci 27(2):365–387 weighted tardiness and weighted earliness of containers. Int J35. Khan, Lutfar R, Sarker R A (2002) An optimal batch size for a Prod Res 37(5):1137–1158 JIT manufacturing system. Comput Ind Eng 42(2–4):127 62. Richard J, Tersine (1994) Principle of inventory and materials36. Kimura O, Terade H (1981) Design and analysis of pull system management. PTR Prentice hall Englood cliffs, New Jersey a method of multistage production control. Int J Prod Res 63. Schonberger R (1987) World Class Manufacturing case Book 19:241–253 implementing JIT & TQC. The Free Press, New York37. Cua K O, Kathleen E, Mekone, Schroeder R G (2001) 64. Rosemary et al (2001) The production performance benefits Relationship between implementation of TQM, JIT and TPM from JIT implementation. J Oper Manage 19:81–96 and manufacturing performance. J Oper Manag 19:675–694 65. Rosemary R, Fullerton, Cheryl S, Mc Watters, Fawson C38. Krieg G, Kuhn H (2002) A decomposition method for multi (2003) An examination of the relationship between JIT and product kanban systems with setup times and lost sales. IIE financial performance. J Oper Manage 20:383–404 Trans 34(7):613 66. Bowden RO, Hall J D, Usher J M (1996) Integration of39. Kulwiec R A (1997) The down side of JIT. Mod Mat Handl 52 evolutionary programming and simulation to optimize a pull (10):3 production system. Comput Ind Eng 31(1&2):217–22040. Kutc S (1997) Optimal buffer allocation strategy for minimiz- 67. De Smet R, Ludo G (1998) Using simulation to evaluate the ing work in processinventory in un paced production lines. IIE introduction of a kanban sub system within an MRP-Controlled Trans 29:81–88 manufacturing environment. Int J Prod Econ 56 & 57:111–12241. Lai CL, Lee WB, W.H.Ip (2003) A study of system dynamics 68. Salameh MK, Ghattas RE (2001) Optimal Just in Time buffer in Just in Time logistics. Journal of Material Processing inventory for regular preventive maintenance. Int J Prod Econ Technology 138:265–269 74:157–16142. Lee CY Lin, Uzosoy CS, Wong R (1994) K., Implementation 69. Ryan S M, Choobineh F (2003) Total WIP and WIP mix for a of demand pull system in a job shop environment. Int J Prod CONWIP controlled jobshop. IIE Trans 35:405–418 Res 32(11):2915–2927
    • 40870. Ryan SM, Bruno B, Choobineh F F (2000) Determining 84. Takashashi K, Nakamura N (2002) Decentralized reactive inventory levels in CONWIP controlled jobshop. IIE Trans kanban system. Eur J Oper Res 139(2):262 32:105–114 85. Tardif V, Maaseidvaag L (2001) An adaptive approach to71. Sarker B R, Balan Chidambaram V (1996) Operations planning controlling kanban systems. Eur J Oper Res 132(2):411 for kanbans between two adjacent workstations. Comput Ind 86. Tayfur A, Goang A S (2000) Pull type manufacturing system Eng 31(1&2):221–224 with multiple product types. IIE Trans 32:115–12472. Sarker Bhaba, Balan Chidambaram V (1999) Operations 87. Turbo days (N) Time since Jan 12 (1996) http://www/geocities. planning for a multi-stage kanban system. Eur J Oper Res com/timesquare/1848/japan21.html 112(2):284–303 88. Uday S, Karmarkar (1987) Lot-sizing and sequencing delays.73. Seki Y, Hoshino N (1999) Transient behavior of a single-stage Manage Sci 33:419–423 kanban system based on the queuing model. Int J Prod Econ 60 89. Dudek Villeda R, Smith RML (1988) Increasing the production & 61:369–374 rate of just in time production system with variable operation74. Shahabudeen P, Krishnaiah K (1999) Design of a Bi-Criteria times. Int J Prod Res 26(11):1747–1768 kanban system using Genetic Algorithm. Int J Manage syst 15 90. Vito A, Dassisti M, Okogbaa G O (1995) Approximation (3):257–274 approach for performance analysis production lines under a75. Shahabudeen P, Gopinath R, Krishnaiah K (2002) Design of bi- kanban discipline. Int J Prod Econ 40:197–207 criteria kanban system using simulated annealing technique. 91. Womack JP et al (1991) The machine that change the world - Comput Ind Eng 41(4):355 the story of lean production. Rawson /Harper perennial, New76. Shahabudeen P, Krishniah K, Thulasinarayanan (2003) Design York of two card dynamic kanban system using a simulated 92. Womack JP, Jones DT (1994) From the lean production to lean annealing algorithm. International Journal Advanced Manufac- enterprise. Harward Business Review 93–103 turing Technology 21:754–759 93. Woolsey R ED, Bowden R O, Hall J D, Hadley W H (1999)77. Sharadapriyadarshini, Rajendran C (1997) Scheduling in Closed-form solution to kanban sizing problem. Prod Invent kanban-controlled flow shop with dual blocking mechanisms Manage J 40(1):1–3 and missing operations for part types. Int J Prod Res 35 94. Xiaobo Z, Ohno K (1997) Algorithms for sequencing mixed (11):3133–3156 models on an assembly line in a JIT Production system.78. Sharadapriyadarshini, Rajendran C (1997) Heuristics for Comput Ind Eng 32(1):47–56 scheduling in kanban system with dual blocking mechanisms. 95. Herer Yale T, Levishalom (2000) The Kanban Assignment Eur J Oper Res 103:439–452 problem- a non integral approach. Eur J Oper Res 120:260–27679. Spearman ML, Woodruff DT, Hopp WJ (1990) CONWIP a pull 96. Yang (2000) Managing a flow line with single-Kanban, dual- alternative to Kanban. Int J Prod Res 28:879–894 Kanban or Conwip. Prod Oper Manage 9:34980. Stefan Minner (2003) Multiple supplier-inventory model in 97. Yannick F, Maria D, Yves D (1995) On the design as supply chain management: a review. Int J Prod Econ generalized kanban control system. Int J Oper Prod Manage 81&82:265–279 15:981. Stockton DJ, Lindley RJ (1995) Implementing kanban with 98. Yavuz IH, Satir A (1995) A kanban based simulation study of a high variety / load volume manufacturing environment. Int J mixed model just in time manufacture line. Int J Prod Res 33 Oper Prod Manage 15(7):47–59 (4):1027–104882. Sugimori Y et al (1977) Toyota production system and kanban 99. Yoichi S, Naoto H (1999) Transient behavior of single stage system - materialization of just in time and respect for human kanban system based on the queuing model. Int J Prod Econ 60 system. Int J Prod Res 15(6):553–564 & 61:369–37483. Swanson C A, Lankford W M (1998) Just In Time 100. Yves D, Liberopoulos G (2000) Extended kanban control manufacturing. Bus Process Manag J 4(4):333 system: combining kanban and base stock. IIE Trans 32: 369–386