Literature review of JIT-KANBAN systemAbstractIn this paper， JIT （Just-In-Time） KANBANliterature survey was carried out and presented.Theintroductory section deals with the philosophy of JIT，andthe concept involved in the push and pull system. Theblocking mechanisms in the kanban system are alsodiscussed elaborately。 Besides these sections, the importanceof measure of performance （MOP） and the applicationof the same with respect to JIT-KANBAN arepresented。 The recent trends in the JIT-KANBAN arediscussed under the heading “Special cases”. In this review，100 stateof-art research papers have been surveyed. Thedirections for the future works are also presented.1 IntroductionJust InTime （JIT) manufacturing system was developedby Taiichi Ohno which is called Japanese “Toyotaproduction system”JIT manufacturing system has theprimary goal of continuously reducing and ultimatelyeliminating all forms of wastes (Brown et al。 5，Ohno54， Sugimori et al. 82)。 Based on this principle,Japanese companies are operating with very low level ofinventory and realizing exceptionally high level of qualityand productivity （Richard J。 Tersine 62, James H。 Greene 30）。 JIT emphasizes “zero concept” which meansachievement of the goals of zero defects， zero queues,zero inventories, zero breakdown and so on. It ensures thesupply of right parts in right quantity in the right place andat the right time. Hence, the old system of materialacquisition and, buyer and seller relationships are changedto new revolutionary concepts （Womack et al。 91,Womack and Jones 92, Markey et al. 45)。 Similarly，JIT becomes an inevitable system at plant level, whichintegrates the cellular manufacturing, flexible manufacturing，computer integrated manufacturing and Robotics(Schonberger 63， Golhar 12）.Due to the technological advancement, the conventionalmethod of push production system linked with MaterialRequirement Planning (MRP） was changed to pull type JITproduction system to meet out the global competition，where the work-inprocess （WIP) can be managed andcontrolled more accurately than the push productionsystem （Mason Paul 46）.KANBAN system is a new philosophy, which plays asignificant role in the JIT production system。 Kanban isbasically a plastic card containing all the informationrequired for productionassembly of a product at each stageand details of its path of completion。 The kanban system isa multistage production scheduling and inventory controlystem。 These cards are used to control production flowsand inventory。 This system facilitates high productionvolume and high capacity utilization with reduced productiontime and workin-process。The objectives of this paper are as listed below1） Critical review of JIT literature。2） Segregating the different research articles of JIT。3) Exploring the recent trends in JITKanban system andderiving directions for future research.In this paper, the articles are reviewed and an appropriateclassification is presented。The kanban study was madeelaborately, since it acts as a basic communicator and feedbackagent to the JIT system. Push and pull system，principle of operation of kanban cards， Blocking mechanism，Toyotas formula， and the measures of performances(MOP) are also discussed in this paper。 The latest trends inJITKanban system are also addressed separately under theheading “Special cases"。 Finally， the directions for futureresearches are presented。6。1.1 Flow shopKanban system is widely implemented in repetitivemanufacturing environment. Fora single card operationalsystem， Sharadhapriyadarishini et al. 77 have developedtwo heuristics and proved that these are more efficient。Saradhapriyadarishini et al。 78 have proposed a recursiveequation for scheduling the single card kanban system withdual blocking. They proposed a heuristic with twinobjectives of minimizing the sum of total weighted timeof containers and weighted flow time of part-types.Rajendran 61 has done a work on two card flow shopscheduling with n parttypes. In this paper, mathematicalmodels for time tabling of containers for different problemshave been formulated。 Then, a heuristic was developed tominimize the sumof weighted flow time， weighted earliness,and weighted tardiness of containers. Hemamalini et al. 22have done similar work。 In this work, the heuristic developedis simulated annealing algorithm. This is compared withrandomsearchmethod. In these papers， the comparisons aredone only based on mean relative percentage increase。Instead of this approach, comparisons based on completeANOVA experiments would provide reliable inference.Peter Brucker et al。 58 have carried out research onflow shop problem with a buffer of limited capacitybetween two adjacent machines. After finishing theprocessing of a job on a machine, either the job is to beprocessed on the following machine or it is to be stored inthe buffer between these machines. If the buffer iscompletely occupied， the job has to wait on its currentmachine but blocks this machine for other jobs。 In thispaper, they determined a feasible schedule to minimize themakespan using tabu search。 The results of the problemusing tabu search were compared with that of benchmarkinstances. The comparisons are done only based on relativeimprovements。 Instead of this approach， comparisonsbased on complete ANOVA experiments would providereliable inference。6。1。2 Assembly lineAssembly 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 achallenging task. Drexl et al. 16 considered an assemblyline sequencing mixed model problem. It is a combinatorialproblem。 They formulated this combinational problem asinteger programming model。 This model can be used onlyfor small size problems due to the limitations of operationsresearch software with respect to handling the number ofvariables and constraints， which are present in the integerprogrammingmodel. Xiaobo et al。 94 have consideredsimilar work on mixed modelassembly line sequencingproblem with conveyor stoppages。 They proposed branchand bound algorithm， and simulated annealing algorithmfor finding the optimal solution and suboptimal solution ofthe mixed-model sequencing problem， respectively tominimize the total conveyor stoppage time。 The branchand-bound method was devoted to find the optimalsolution of small-sized problems， whereas the simulatedannealing method was used to cope with largescaleproblems to obtain a good suboptimal solution。 Future，research on simulated annealing applied to this problemcan be directed to establish a better seed generationalgorithm。 However， the practitioner should spend considerabletime in fixing the parameter called temperature(T） in the simulated annealing algorithm by trail and errormethod before actually solving the problem。6.1.3 Batch production systemIn a batch production system, the switching over from oneproduct to other product depends on many factors such asstock reaching to the threshold level, different priorityschemes， economical setups, etc。 Tafur Altiok et al。 86have dealt this issue differently for the pull typemanufacturing system with multi product types。 In thispaper， they developed an iterative procedure to approximatelycompute the average inventory level of eachproduct as finished goods using different priority schemes.In this paper, the demand arrival process is assumed to bea poisson distribution and processing times and the set-uptimes are arbitrarily distributed. But, in practice， theprocessing times may follow other distributions, viz.,normal, uniform, exponential， etc。 which are not experimentedin this paper。 Khan et al。 35 addressed theproblem of manufacturing system that procures rawmaterials from vendors in lot and convert them intofinished products。 They estimated production batch sizesfor JIT delivery system and designed a JIT raw materialsupply system. A simple algorithm was developed tocompute the batch sizes for both manufacturing and rawmaterial purchasing policies.7 JIT integration, implementation and benefitsJustin-time is a manufacturing philosophy by which anorganization seeks continuous improvements. For ensuringcontinuous improvements， it is necessary for any organizationto implement and integrate the JIT and JIT relatedareas。 If it is practiced in its true sense， the manufacturingperformance and the financial performance of the systemwill definitely improve.Swanson et al. 83 have reiterated that proper planningis essential for implementation of a JIT manufacturingsystem and a commitment from top management is a prerequisite.Cost benefit analysis is to be studied initially withthe knowledge of key items such as the cost of conversionto a JIT system and time period of conversion. Cook et al.11， in their case study for applying JIT in the continuousprocess industry， show improvements in demand forecastand decrease in leadtime variability。The relationship between implementation of TQM, TPMand JIT will lead to improvement in the manufacturingperformance (Kribty et al。 37）。 Further Huang 23discusses the importance of considering the integration ofTPM， JIT， Quality control and FA (Factory Automization）.Imai 27 believes that TQM and TPM are the two pillarssupporting the JIT production system. Kakuro Amasaka 32 proposes a new JIT management system， which helpsto transfer the management technology into managementstrategy。Fullerton et al. 65 have conducted a study in 253 firmsin USA to evaluate empirically whether the degree withwhich a firm implements the JIT practices affects the firmsfinancial performance。 From their study， JIT manufacturingsystem will reap sustainable rewards as measured byimproved financial performance。 Also， they studied thebenefits of JIT implementation in 95 firms in USA. Theyhave concluded that JIT implementation improves theperformance of the system， because of resultant quality benefits， time based benefits, employees flexibility, accounting simplification， firms profitability and reducedinventory level。8 ConclusionThe growing global competition forces many companies toreduce the costs of their inputs so that the companies canhave greater profit margin。 There are considerableadvancements in technology and solution procedures inreality, to achieve the goal of minimizing the costs ofinputs。 JIT-KANBAN is an important system， which isused in production lines of many industries to minimizeworkinprocess and throughput time, and maximize lineefficiency。 In this paper, the authors have made an attemptto review the state-ofart of the research articles in the area “JITKANBAN system”. After a brief introduction to pushand pull systems， different types of kanban and theiroperating principles， blocking mechanisms, the authorshave classified the research articles under JIT-KANBANsystem into five major headings, viz。, empirical theory,modeling approach， variability and its effect， CONWIP andJIT-SCM. Also， the authors have provided a section forspecial cases under JIT-KANBAN. This paper would helpthe researchers to update themselves about the currentdirections and different issues under JIT-KANBANsystem， which would further guide them for their futureresearches.The directions for future researches are presented below。The flow shop as well as mixed model assembly lineproblems come under combinatorial category。 Hence，metaheuristics viz。, simulated annealing, genetic algorithmand tabu search may be used to find solution todetermine the minimum number of kanbans and othermeasures. In simulated annealing algorithm， researcherscan aim to device a better seed generation algorithm whichwill ensures better starting solution. In most of the papers,comparisons are done only based on relative improvements. Instead of this approach, comparisons based oncomplete ANOVA experiments would provide reliableinferences.This algorithm developed by Elizabeth Vergara et al.18 uses only twopoint crossover genetic operators. Athird genetic operator may be introduced to further improvethe performance of the evolutionary algorithm. Theevolutionary algorithm may be modified to handle complexsupply chain problem. In JIT-SCM related researchworks, effort should be directed to develop simulation aswell as metaheuristics to derive results under probabilisticconditions。In the work of Sarah M. Rayan et al. 69, the applicationof single chain analysis for multiple chain operation raisesan open question whether a single WIP level should bemaintained for all products or individual levels for eachproduct。 Further， most of the studies use simulation. Hence，future research shall be directed to develop improvedsearch procedures for finding WIP levels in kanbansystems. As an extension to the work of Krieg et al. 38,a decomposition algorithm can be developed for multiproductkanban systems with state dependent setups。 Theadaptive approach suggested by Tardif et al. 85 may beextended for multistage， multiproduct kanban system。The work of Lai et al。 41 can be extended by includingmore variables and elements and conducting experimentsto investigate the stability of the system under variousconditions such as the sudden increase in demand andrandom demand， experimenting on the system behaviour ofdifferent types of customer and modes of manufacturing.The nested partitioned method provided by Leyuan Shi andShuli Men 43 can be enhanced by incorporating any oneor a combination of the many other heuristics viz.,elaborate partitioning， sampling， backtracking scheme，simulation， etc. Then， they can be applied to combinatorialproblems of this typeAnts colony optimization algorithm is a recent inclusionto the existing meta-heuristics viz。, simulated annealingalgorithm， genetic algorithm and tabu search. So, aresearcher can study the solution accuracy as well asrequired computational time of this algorithm for his/herJIT problem of interest， which falls under combinatorialcategory and compare its results with the results of theother three heuristics (meta-heuristics）。Source：C.Sendil Kumar, R。Panneerselvam， 2007.“Literature review of JITKANBAN system".The International Journal of Advanced Manufacturing Technology， vol.32，no。5,August.pp。393408.中文翻譯JIT看板系統(tǒng)的文獻回顧摘要：在本文中，通過對JIT（實時）看板文獻的調(diào)查提交相關(guān)的報告。報告介紹部分JIT處理的理念以及概念包含了推式及拉式系統(tǒng)這些內(nèi)容，同時，對看板系統(tǒng)中的阻塞機制也進行精心地討論。在后面各節(jié)中，介紹JIT看板的應(yīng)用程序和性能（MOP)在實施中的重要性.此外,文章也討論了JIT最近出現(xiàn)的“特殊情況"的現(xiàn)象。最后，本文通過對100個先進國家的研究論文進行調(diào)查，提出了未來的發(fā)展方向。1 介紹被日本稱為實時 （JIT） 制造系統(tǒng)的“豐田生產(chǎn)方式”由大野耐一開發(fā)的。JIT制造系統(tǒng)是以不斷減少和最終清除各種各樣的廢物為主要目標(biāo)(布朗等人 5，大野54，蘇格莫里等人82）?；谶@一原則,日本企業(yè)保持非常低的庫存,實現(xiàn)極高的質(zhì)量和生產(chǎn)率（理查德特斯恩62,詹姆斯H格林30）的經(jīng)營水平。JIT強調(diào)“零概念”，這意味著零缺點、零隊列、零庫存、零故障等目標(biāo)需要實現(xiàn)。它可以在適當(dāng)?shù)奈恢煤瓦m當(dāng)?shù)臅r間,確保供應(yīng)正確數(shù)量的部件。在此基礎(chǔ)上，把重大收購，舊系統(tǒng)更改以及買方和賣方的關(guān)系作為新的革命理念來研究(沃馬克等，沃馬克91，和瓊斯92，馬凱等人45）。同樣,JIT在工廠運營中成為不可避免的制造系統(tǒng)，該系統(tǒng)擁有單元制造,柔性制造，計算機集成制造和機器制造（舍恩伯格63，格蘭翰12）.技術(shù)的進步以及滿足全球競爭的需要,將傳統(tǒng)的推動生產(chǎn)系統(tǒng)與物料需求計劃(MRP）結(jié)合，形成拉式JIT生產(chǎn)系統(tǒng).在那種情況下，可以對拉式制造系統(tǒng)生產(chǎn)的制品（WIP）進行良好的管理和控制（梅森保羅46）?？窗逑到y(tǒng)是一個新的理念，在JIT生產(chǎn)系統(tǒng)中起著重要作用.看板基本上包含在每個生產(chǎn)階段,是一種在完成裝配中需要很多細(xì)節(jié)信息的塑料卡。同時，看板系統(tǒng)也是一個需要多階段生產(chǎn)計劃和庫存控制的系統(tǒng)。這些塑料卡用來控制生產(chǎn)過程中多余的庫存.系統(tǒng)看板的使用可以提高產(chǎn)量和利用率，減少在工作過程中的生產(chǎn)時間。本文的目標(biāo)是如下所示：（1）嚴(yán)格審查JIT文獻(2)在不同文章中，研究JIT的區(qū)別（3）探討JIT看板系統(tǒng)的最新趨勢和未來研究方向本文在精心審查和研究的基礎(chǔ)上對提出的看板進行適當(dāng)?shù)姆诸?，因為作為一種在代理人使用它后需要進行反饋的JIT系統(tǒng)。本文還討論看板卡，豐田公式,攔截機制，如何實施看板應(yīng)用程序和性能（MOP）的措施和推拉系統(tǒng)的工作原理，也包括文章中JIT看板系統(tǒng)最新“特殊情況”的處理.最后，對未來的研究提出了方向。6.1.1 流水作業(yè)看板系統(tǒng)是在重復(fù)生產(chǎn)環(huán)境中廣泛實施。單卡看板業(yè)務(wù)系統(tǒng)可以分成兩個被證明具有很高效率的啟發(fā)式（薩拉德哈勃利亞達(dá)等人77）.同時，他們提出單卡看板業(yè)務(wù)系統(tǒng)與雙阻塞的遞歸方程（薩拉德哈勃利亞達(dá)等人78），還提出利用啟發(fā)式部分加權(quán)流量時間的總和計算最小化總加權(quán)時間。然而，他在兩個單卡流水車間做了N個啟發(fā)式部分類型的調(diào)度工作（拉基杰萊61）。本文針對不同的問題，例如以盡量減少計算在加權(quán)時間早到和加權(quán)時間遲到這兩種模型下的加權(quán)總和為目標(biāo)而開發(fā)了一種啟發(fā)式算法,提交不同的數(shù)學(xué)模型報告。另外，他們也做了類似的工作（赫馬馬利尼等人22）。在這項工作中，開發(fā)的啟發(fā)式算法被稱為模擬退火算法。通過與其他算法的比較,在相對的基礎(chǔ)上只能得到平均增幅，而不是基于在實驗的基礎(chǔ)上用完整的方差比較分析得出可靠的推論.他們對兩個相鄰的計算機之間關(guān)于容量有限的緩沖區(qū)流水作業(yè)問題進行了研究（彼得布魯克等人58）。在計算機上完成作業(yè)的處理后,接著，要處理在指定計算機之間的緩沖區(qū)中存儲的任務(wù)。如果緩沖區(qū)被完全占用,那當(dāng)前計算機上作業(yè)只有在阻止這臺機器的其它工作情況下實施。在本論文中，他們已確定一個可行的時間表，用來盡量減少在最大完工時間里使用禁忌搜索.使用禁忌搜索產(chǎn)生的問題可以通過比較基準(zhǔn)來反映結(jié)果。該比較基準(zhǔn)只是基于相對改善，而不是在實驗的基礎(chǔ)上通過完整的方差比較分析得到可靠的推論.6。1.2 裝配線商店在經(jīng)營某些部分時進行類似流動裝配線組裝的排序。在有多產(chǎn)品組裝線情況下，排序工作是一項艱巨的任務(wù).它被認(rèn)為是一種混合模式裝配線排序問題，也是一個組合的問題，他們制定了這個整數(shù)規(guī)劃模型組合解決相應(yīng)的問題（蒂萊爾等16)。該模型只可用于解決小規(guī)模問題，是因為有相關(guān)處理變量的限制，這也是目前運籌學(xué)軟件受到整數(shù)規(guī)劃模型組合的限制。然而,他們已經(jīng)考慮到類似混流裝配線排序有輸送時停頓問題現(xiàn)象出現(xiàn)，提出了尋求最優(yōu)解和次優(yōu)的混合模式排序問題的解決方案，在總補時階段分別采用分支和模擬退火算法使得排序最小化（劉曉波等人94）。分支定界算法是專門尋找最優(yōu)解解決小型問題，而模擬退火法處理規(guī)模大的問題取得了良好的次優(yōu)的解決方案。今后,規(guī)模大的問題適用于模擬退火研究,同時，還可以建立一個更好的遺傳算法。不過，實際上在確定之前，學(xué)者應(yīng)該花費相當(dāng)長的時間對模擬退火算法錯誤方法進行跟蹤并且得出解決參數(shù)調(diào)用限度的問題.6。1.3 批量生產(chǎn)系統(tǒng)在一個批次的生產(chǎn)體系中，從一種產(chǎn)品轉(zhuǎn)換到其他產(chǎn)品依賴實現(xiàn)增值的水平,不同的優(yōu)先配置計劃,合理的經(jīng)濟設(shè)置等許多因素。他認(rèn)為拉式制造系統(tǒng)可以處理產(chǎn)品多種不同類型的問題（塔夫艾提克等86）.在本論文中，他們制定了使用一個不同的迭代過程來計算近似每個產(chǎn)成品庫存平均水平的優(yōu)先配置計劃。還要求被假設(shè)處理的時間和設(shè)置時間是任意泊松分布的。但是，在實踐中,處理時間可能跟隨其他的分布，如正常，均勻，指數(shù)等，這些不屬于本文研究.他們討論了關(guān)于不少廠商從采購原料，并轉(zhuǎn)換成成品的產(chǎn)品這過程中的制造系統(tǒng)問題，估計JIT在配送系統(tǒng)中生產(chǎn)批量的大小,設(shè)計了JIT的原料供應(yīng)體系（漢等35）。就這樣一個簡單的算法被開發(fā)用來計算原材料制造和采購政策的批量大小.7 JIT的集成、實施和效益實時是一種制造理念，需要一個公司阻礙不同的時間不斷改進。為了保證不斷改善,有必要對任何組織和相關(guān)領(lǐng)域?qū)嵤㎎IT和JIT的整合。如果從檢驗真正實踐意義上講，JIT系統(tǒng)的生產(chǎn)性能和財務(wù)表現(xiàn)一定會改善。他們重申,適當(dāng)規(guī)劃實施的JIT生產(chǎn)系統(tǒng)和一個由最高管理層的承諾是必不可少的先決條件（斯萬森等人83）。然而，成本效益分析是進行研究企業(yè)財務(wù)重點項目的初步知識，就像JIT系統(tǒng)的成本和時間是JIT實施的基礎(chǔ)。流水作業(yè)型的企業(yè)實施JIT，需要改善預(yù)測時間，提高時間價值（庫克等人11）。TPM和JIT之間實現(xiàn)全面質(zhì)量管理關(guān)系,將導(dǎo)致生產(chǎn)性能的改善（凱利比等37）.此外,他討論了TPM，準(zhǔn)時生產(chǎn),質(zhì)量控制和FA（工廠自動化）一體化的重要性（黃23)。另外，他認(rèn)為TQM和TPM是實施JIT生產(chǎn)系統(tǒng)的兩個支柱(今井27）.有人提出了一種新的JIT管理系統(tǒng)，這有助于管理策略轉(zhuǎn)移到管理技術(shù)(卡庫羅艾瑪薩克32）。他們在美國進行了一項研究，即從253家公司的財務(wù)經(jīng)營業(yè)績的好壞程度，評價實施JIT做法影響公司的興衰(富勒頓等人65）.他們研究用改善財務(wù)的表現(xiàn)衡量JIT可持續(xù)在生產(chǎn)系統(tǒng)中獲得回報。此外，還研究了美國95家公司實施JIT得到好處.他們的結(jié)論是實施JIT基于時間合理利用和員工的靈活性,提高該系統(tǒng)的性能質(zhì)量和降低庫存水平，由此產(chǎn)生的盈利能力提高。8 結(jié)論不斷的全球競爭迫使許多公司減少其投入的成本，這樣的公司可以有更大的利潤空間。在現(xiàn)實技術(shù)的進步和解決方案增多的情況下，公司把達(dá)到最大限度地降低投入成本作為目標(biāo)。JIT看板是一項在許多產(chǎn)業(yè)生產(chǎn)線中起到重要作用的系統(tǒng)，通過盡量減少工作中的進程和吞吐量的時間，最大限度地提高生產(chǎn)線效率。在本文中，作者嘗試在研究先進文章的同時去探索JIT看板系統(tǒng)在該領(lǐng)域使用的先進設(shè)備。在簡要介紹推拉系統(tǒng)，不同看板類型，阻斷機制及其工作原理后，作者將JIT看板系統(tǒng)的研究文章分為五個大內(nèi)容：即實證理論，建模方法，多變性和其效果，競爭的在制品和JIT供應(yīng)鏈管理。此外，作者們提供了一個JIT模式下，出現(xiàn)部分看板的特殊情況.本文將幫助研究人員對不同問題下的JIT看板系統(tǒng)提供目前的更新發(fā)展方向，這將進一步引導(dǎo)他們未來研究的方向。對未來的研究方向介紹如下：啟發(fā)式算法，模擬退火算法,遺傳算法和禁忌搜索可能被用來尋找解決像流水車間混合裝配屬于組合這類問題的最優(yōu)方案。當(dāng)需要使用模擬退火算法時,研究人員也可以考慮使用遺傳算法，這將能保證解決問題需要更好的方案。在大部份的文件中，相互間的比較僅是基于相對改善。而不是在實驗的基礎(chǔ)上通過完整的方差比較分析得到可靠的推論。他們開發(fā)了這個算法，僅使用了兩個點的交叉遺傳算法（伊麗莎白,貝爾加拉等人18）。后來可能引入遺傳操作,進一步提高了遺傳算法的性能。遺傳算法在處理復(fù)雜的供應(yīng)鏈問題可能被修改.在JIT供應(yīng)鏈管理相關(guān)的研究工作中，應(yīng)該努力在啟發(fā)式概率條件下得出開發(fā)仿真的結(jié)果。關(guān)于單鏈分析多個連鎖經(jīng)營提出了一個懸而未決的問題，它是一個由單一制品水平為所有產(chǎn)品提供單位產(chǎn)品水平的問題（薩拉米娃蒂等69)。此外，大多數(shù)研究利用電腦模擬。因此，為了研究的發(fā)展，今后的研究方向應(yīng)是發(fā)展以及改善看板系統(tǒng)在制品搜索程序中的水平（克里格等38）。一個分解算法可開發(fā)多種符合計算國家相關(guān)的設(shè)置看板系統(tǒng)的產(chǎn)品的方法。該合理方法是由他所提出的(塔蒂夫等85）。該方法可適用于擴展到多產(chǎn)品多級看板系統(tǒng)。有人為了研究該算法的延伸性進行并實驗，包括更多的變數(shù)和因素，如在需求和隨機需求的突然增加情況下以及在各種條件下，對不同類型的客戶系統(tǒng)使用生產(chǎn)模式的穩(wěn)定性進行行為試驗(勵等41）。嵌套分區(qū)方法是由他們提供的,可以精心劃分納入任何一個或其它的許多如抽樣，回溯方案,仿真等增強啟發(fā)式的組合(樂苑世與趙樹理男士43)。然后，他們可以應(yīng)用這種方法解決這種類型的組合問題。螞蟻的蟻群優(yōu)化算法是最近加入到現(xiàn)有的啟發(fā)式,模擬退火算法，遺傳算法和禁忌搜索中去的。因此，研究人員可以根據(jù)自己的興趣和所需的計算時間針對JTI問題的組合分別采取不同的算法，研究其他三種啟發(fā)式(啟發(fā)式）的結(jié)果并比較結(jié)果的準(zhǔn)確性。出處：印度 C。塞蒂庫馬爾，R。帕爾塞沃瑪，JIT看板系統(tǒng)的文獻回顧,先進制造技術(shù)國際雜志，第32卷（5)，2007(8）：393-408。翻譯部分為摘要、第一、六、七節(jié)。