Academic literature on the topic 'Job Shop Scheduling Problem (JSSP)'

This thesis aims to review and analyze the scheduling problem in general and Job Shop Scheduling Problem (JSSP) in particular and the solution techniques applied to these problems. The JSSP is the most general and popular hard combinational optimization problem in manufacturing systems. For the past sixty years, an enormous amount of research has been carried out to solve these problems. The literature review showed the inherent shortcomings of solutions to scheduling problems. This has directed researchers to develop hybrid approaches, as no single technique for scheduling has yet been successful in providing optimal solutions to these difficult problems, with much potential for improvements in the existing techniques. The hybrid approach complements and compensates for the limitations of each individual solution technique for better performance and improves results in solving both static and dynamic production scheduling environments. Over the past years, hybrid approaches have generally outperformed simple Genetic Algorithms (GAs). Therefore, two novel priority heuristic rules are developed: Index Based Heuristic and Hybrid Heuristic. These rules are applied to benchmark JSSP and compared with popular traditional rules. The results show that these new heuristic rules have outperformed the traditional heuristic rules over a wide range of benchmark JSSPs. Furthermore, a hybrid GA is developed as an alternate scheduling approach. The hybrid GA uses the novel heuristic rules in its key steps. The hybrid GA is applied to benchmark JSSPs. The hybrid GA is also tested on benchmark flow shop scheduling problems and industrial case studies. The hybrid GA successfully found solutions to JSSPs and is not problem dependent. The hybrid GA performance across the case studies has proved that the developed scheduling model can be applied to any real-world scheduling problem for achieving optimal or near-optimal solutions. This shows the effectiveness of the hybrid GA in real-world scheduling problems. In conclusion, all the research objectives are achieved. Finaly, the future work for the developed heuristic rules and the hybrid GA are discussed and recommendations are made on the basis of the results.

Mestrado em Decisão Económica e Empresarial<br>No processo produtivo de uma indústria de fitofarmacêuticos deparamo-nos com problemas no sequenciamento e afetação de tarefas necessárias a encomendas. Este estudo destina-se à fábrica da maior multinacional portuguesa no ramo dos agro-químicos. A SAPEC Agro aposta numa estratégia vencedora no negócio agrícola e, neste ramo, a competição não é entre empresas, mas sim entre as cadeias de abastecimento. Com o crescimento da empresa e a sua internacionalização, inserida num ambiente altamente competitivo, torna-se fundamental automatizar o processo de planeamento da produção, que até aos dias de hoje tem sido feito por um engenheiro industrial. O presente estudo tem como objetivo encontrar um sequenciamento de tarefas (scheduling), ou seja, determinar uma afetação ótima das tarefas às máquinas, minimizando o tempo de execução total (makespan). Este tipo de problema é um clássico da literatura e é conhecido como um problema de job-shop scheduling com uma sequência dependente de tempos de setup (JSP-SDST). Um JSP-SDST apresenta, uma complexidade NP-difícil. Sendo um problema de difícil otimização, constituiu, assim, um desfio nas áreas da Investigação Operacional e Ciência da Computação. A abordagem escolhida passou primeiramente por definir o espaço de soluções admissíveis (S.A.) e neste encontrar uma solução ótima (sob determinadas condições), através do algoritmo Branch-and-Bound, recorrendo-se particularmente ao tipo de pesquisa Depth First Search (B&B-DFS) no espaço das soluções admissíveis. Neste estudo são apresentados outros tipos de pesquisa, também baseados em B&B, por forma a avaliá-los. Foi desenvolvida uma interface gráfica centrada no utilizador para que este possa usufruir do presente estudo, sem ter de lidar com a complexidade envolvida. A interface gráfica implementada é viável e, em articulação com os procedimentos de otimização escolhidos, constitui uma mais-valia para a organização, esperando-se que possa vir a ser instrumento de trabalho futuro.<br>The production process in the plant protection industry, as in many other industries, faces very often problems on how to schedule tasks that need to be completed in order to respond to a given set of orders. This study is made for the largest Portuguese multinational company in the agro-chemicals business, SAPEC Agro. SAPEC Agro relies on a winning strategy on this business, a business where competition is made between supply chains instead of companies. To follow the company’s growth and its continuous internationalization process, inserted in a highly competitive environment, this project aims to maximize the automation and optimize the production planning, which is now made by an Industrial Engineer. This study’s main goal is to develop an algorithm that will be able to find the optimal solution (under certain constraints) for the scheduling of tasks in the production process. That means determining the best possible sequence of tasks to each machine in order to minimize the total makespan. This type of problem is classic in the literature, and it is known as a Job-Shop Scheduling Problem with Sequence Dependent Setup Times (JSP-SDST). The resolution of the JSP-SDST is difficult, as in terms of computational complexity it is considered an NP-Hard problem, thus being a challenge in the areas of Operations Research and Computer Science. As the problem is applied in an industrial environment, the scheduling is useful if the algorithms respond in a reasonable amount of time, allowing the production managers to get real-time support when decisions need to be taken. The chosen approach was first to define an admissible solution space (some constraints to the allocations were applied), and then to find the optimum through a Branch-and-Bound method which uses a Depth-First-Search as method of search in the solution space. Other search methods and heuristics, also based on Branch-and-Bound are applied as well, in order to meet the time complexity constraints. A graphical interface is developed allowing its use even by those unfamiliar with the complexity of the problem. Users need just to include the inputs (orders and quantities of each product) and the program generates a schedule for the input orders. This work’s main goal is the development of a program that would be useful and would add value to the organization in study, the SAPEC Agro.

We define a job-shop scheduling problem with three dynamic decisions: assigning duedates to exogenously arriving jobs, releasing jobs from a backlog to the shop floor, and sequencing jobs at each workstation in the shop. The objective is to minimize both the work-in-process (WIP) inventory on the shop floor and the due-date lead time (due-date minus arrival date) of jobs, subject to an upper bound constraint on the proportion of tardy jobs. A general two-step approach to this problem is proposed: (1) release and sequence jobs in order to minimize the WIP inventory subject to completing jobs at a specified rate, and (2) given the policies in (1), set due-dates that will attempt to minimize the due-date lead time, subject to the job tardiness constraint. A simulation study shows that this approach easily outperforms other combinations of traditional due-date setting,job release, and priority sequencing policies. As a result of the study, three scheduling principles are proposed that can significantly improve the performance of a job shop. In particular, better due-date performance can be achieved by ignoring due-dates on the shop floor.

Mestrado em Engenharia e Gestão Industrial<br>Este trabalho aborda um novo tipo de problema de escalonamento que pode ser encontrado em várias aplicações do mundo-real, principalmente na indústria transformadora. Em relação à configuração do shop floor, o problema pode ser classificado como flexible job-shop, onde os trabalhos podem ter diferentes rotas ao longo dos recursos e as suas operações têm um conjunto de recursos onde podem ser realizadas. Outras características de processamento abordadas são: datas possíveis de início, restrições de precedência (entre operações de um mesmo trabalho ou entre diferentes trabalhos), capacidade dos recursos (incluindo paragens, alterações na capacidade e capacidade infinita) e tempos de setup (que podem ser dependentes ou independentes da sequência). O objetivo é minimizar o número total de trabalhos atrasados. Para resolver o novo problema de escalonamento proposto um modelo de programação linear inteira mista é apresentado e novas abordagens heurísticas são propostas. Duas heurísticas construtivas, cinco heurísticas de melhoramento e duas metaheurísticas são propostas. As heurísticas construtivas são baseadas em regras de ordenação simples, onde as principais diferenças entre elas dizem respeito às regras de ordenação utilizadas e à forma de atribuir os recursos às operações. Os métodos são designados de job-by-job (JBJ), operation-by-operation (OBO) e resource-by-resource (RBR). Dentro das heurísticas de melhoramento, a reassign e a external exchange visam alterar a atribuição dos recursos, a internal exchange e a swap pretendem alterar a sequência de operações e a reinsert-reassign é focada em mudar, simultaneamente, ambas as partes. Algumas das heurísticas propostas são usadas em metaheurísticas, nomeadamente a greedy randomized adaptive search procedure (GRASP) e a iterated local search (ILS). Para avaliar estas abordagens, é proposto um novo conjunto de instâncias adaptadas de problemas de escalonamento gerais do tipo flexible job-shop. De todos os métodos, o que apresenta os melhores resultados é o ILS-OBO obtendo melhores valores médios de gaps em tempos médios inferiores a 3 minutos.<br>This work addresses a new type of scheduling problem which can be found in several real-world applications, mostly in manufacturing. Regarding shop floor configuration, the problem can be classified as flexible job-shop, where jobs can have different routes passing through resources and their operations have a set of eligible resources in which they can be performed. The processing characteristics addressed are release dates, precedence constraints (either between operations of the same job or between different jobs), resources capacity (including downtimes, changes in capacity, and infinite capacity), and setup times, which can be sequence-dependent or sequence-independent. The objective is to minimise the total number of tardy jobs. To tackle the newly proposed flexible job-shop scheduling problem (FJSP), a mixed integer linear programming model (MILP) is presented and new heuristic approaches are put forward. Three constructive heuristics, five improvement heuristics, and two metaheuristics are proposed. The constructive heuristics are based on simple dispatching rules, where the main differences among them concern the used dispatching rules and the way resources are assigned. The methods are named job-by-job (JBJ), operation-by-operation (OBO) and resource-by-resource (RBR). Within improvement heuristics, reassign and external exchange aim to change the resources assignment, internal exchange and swap intend changing the operations sequence, and reinsert-reassign is focused in simultaneously changing both parts. Some of the proposed heuristics are used within metaheuristic frameworks, namely greedy randomized adaptive search procedure (GRASP) and iterative local search (ILS). In order to evaluate these approaches, a new set of benchmark instances adapted from the general FJSP is proposed. Out of all methods, the one which shows the best average results is ILS-OBO obtaining the best average gap values in average times lower than 3 minutes.

In this work, we investigate job-shop problems where limited capacity buffers to store jobs in non-processing periods are present. In such a problem setting, after finishing processing on a machine, a job either directly has to be processed on the following machine or it has to be stored in a prespecified buffer. If the buffer is completely occupied the job may wait on its current machine but blocks this machine for other jobs. Besides a general buffer model,also specific configurations are considered.The key issue to develop fast heuristics for the job-shop problem with buffers is to find a compact representation of solutions. In contrast to the classical job-shop problem,where a solution may be given by the sequences of the jobs on the machines, now also the buffers have to be incorporated in the solution representation. In this work, we propose two solution representations for the job-shop problem with buffers. Furthermore, we investigate whether the given solution representations can be simplified for specific buffer configurations. For the general buffer configuration it is shown that an incorporation of the buffers in the solution representation is necessary, whereas for specific buffer configurations possible simplifications are presented. Based on the given solution representations we develop local search heuristics in the second part of this work. Therefore, the well-known block approach for the classical job-shop problem is generalized to the job-shop problem with specific buffer configurations.

Flexible job-shop scheduling problem (FJSP) is a generalization of the classical job-shop scheduling problem (JSP). It takes shape when alternative production routing is allowed in the classical job-shop. However, production scheduling becomes very complex as the number of jobs, operations, parts and machines increases. Until recently, scheduling problems were studied assuming that all of the problem parameters are known beforehand. However, such assumption does not reflect the reality as accidents and unforeseen incidents happen in real manufacturing systems. Thus, an optimal schedule that is produced based on deterministic measures may result in a degraded system performance when released to the job-shop. For this reason more emphasis is put towards producing schedules that can handle uncertainties caused by random disruptions. The current research work addresses solving the deterministic FJSP using evolutionary algorithm and then modifying that method so that robust and/or stable schedules for the FJSP with the presence of disruptions are obtained. Evolutionary computation is used to develop a hybridized genetic algorithm (hGA) specifically designed for the deterministic FJSP. Its performance is evaluated by comparison to performances of previous approaches with the aid of an extensive computational study on 184 benchmark problems with the objective of minimizing the makespan. After that, the previously developed hGA is modified to find schedules that are quality robust and/or stable in face of random machine breakdowns. Consequently, a two-stage hGA is proposed to generate the predictive schedule. Furthermore, the effectiveness of the proposed method is compared against three other methods; two are taken from literature and the third is a combination of the former two methods. Subsequently, the hGA is modified to consider FJSP when processing times of some operations are represented by or subjected to small-to-medium uncertainty. The work compares two genetic approaches to obtain predictive schedule, an approach based on expected processing times and an approach based on sampling technique. To determine the performance of the predictive schedules obtained by both approaches with respect to two types of robustness, an experimental study and Analysis of Variance (ANOVA) are conducted on a number of benchmark problems.

Orientador: Pedro Sergio de Souza<br>Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Matematica, Estatistica e Computação Cientifica<br>Made available in DSpace on 2018-07-20T09:57:07Z (GMT). No. of bitstreams: 1 Cavalcante_VictorFernandes_M.pdf: 2124078 bytes, checksum: 931a0a025284734a0bdaf335669442c5 (MD5) Previous issue date: 1995<br>Resumo: Times Assíncronos consistem numa nova técnica para solução aproximada de problemas que tem sido aplicada com sucesso a problemas de Otimização Combinatória. Esta técnica faz uso de diversos algoritmos heurísticos que cooperam entre si e conseguem encontrar soluções que não seriam encontradas pelos mesmos algoritmos quando executados isoladamente. Este trabalho tem como objetivo averiguar a adequabilidade de Tunes Assíncronos como metodolqgia para solução do problema de escalonamento de tarefas conhecido por Job Shop Scheduling Problem (JSP). Este problema é considerado um dos mais complexos dentro da Otimização Combinatória e tem recebido crescente atenção nas últimas décadas devido, principalmante, à sua aplicabilidade a processos industriais. Especificamente, o cerne do presente trabalho foi a elaboração de TImes A"síncronos centrados fundamentalmente em heurísticas de construção para oJob Shop Scheduling Problem. Foram concebidas e testadas novas heurísticas para o ISP e novos fluxos de dados que podem ser facilmente acoplados à arquitetura de um Time Assíncrono. Os Times Assíncronos desenvolvidos foram submetidos a diversas instâncias do JSP. Os bons resultados obtidos, não somente atestaram a viabilidade da nova técnica como ferramenta para solução do ISP, como revelaram a competitividade destes resultados com aqueles produzidos por outros métodos aproximados para o problema.<br>Abstract: Asynchronous Teams (or A-Teams) are a new problem resolution technique that has been succesfully applied to Combinatorial Optimization problems. This technique uses several heurisJic algorithms that cooperate simultaneously with each other and find solutions that would not be found through isolated algorithms. The objective of this work is to verify the suitability of Asynchronous Teams methodology solving the combinatorial problem known by Job Shop Scheduling Problem (JSP). This problem has been appointed as one of the most complex problem of Combinatorial Optimization and has been received special attention due to your industrial applicability. Specifically, the kemel of this work was the implementation of A-Teams based on construction heuristics for the Job Shop Scheduling Problem. New heurisncs for the JSP were developed and new data flows that can be easily incorporated in an A-Team architecture were elaborated.. Several JSP instances were used to test the A-Teams developed.. The good results obtained by these A-Teams not onIy showed the feasibility of such technique solving the JSP, but also revealed that this results are competitive with others one obtained by good aproximated approachs for the JSP.<br>Mestrado<br>Mestre em Ciência da Computação

Orientadores: Pedro Sergio de Souza, Marcus Vinicius Poggi de Aragão<br>Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação<br>Made available in DSpace on 2018-07-22T21:33:18Z (GMT). No. of bitstreams: 1 Haddad_ElaineGaspareto_M.pdf: 1752527 bytes, checksum: 6ecf36f9a07b62f0718f185c0fd4ca43 (MD5) Previous issue date: 1997<br>Resumo: Este trabalho aborda o problema de seqüenciamento de tarefas conhecido como Job Shop Scheduling Problem (JSP). O objetivo aqui é mostrar a adequação de uma técnica conhecida como Times Assíncronos (A-Teams), para resolver este problema de otimização combinatória, que é bastante freqüente em ambientes industriais. Esta abordagem tem sido aplicada com sucesso na resolução de outros problemas, como o Traveling Salesman Problem, o Flow-Shop Problem e até mesmo o próprio Job Shop Problem sob uma abordagem de heurísticas de construção. Esta técnica está baseada na cooperação de algoritmos heurísticos no sentido de obter soluções, possivelmente, melhores que aquelas obtidas quando os mesmos algoritmos são executados isoladamente. Neste trabalho, o enfoque é dado a heurísticas de melhoria. Outros tipos de algorit­mos foram desenvolvidos para compor os A-Teams. Estes A-Teams desenvolvidos foram acoplados a um outro já existente, baseado em heurísticas de construção. Algumas instâncias de JSP foram testadas e os resultados obtidos atestam a adequação desta técnica para a resolução deste problema.<br>Abstract: This work treats the sequencing of tasks problem known as Job Shop Scheduling Problem. The goal here is to show the adequability of a technique known as Assynchronous Teams (A-Teams) to solve this optimization problem which is used in industrial environments. This approach has been applied successfully in the solving of other problems such as the Traveling Salesman Problem, Flow Shop Problem and the Job Shop Problem itself using construction heuristics algorithms. This technique is based on the cooperation of some heuristics algorithms in order to obtain solutions, possibly better then ones obtained when same algorithms are working alone. In this work, the focus is on the development of improvement heuristics algorithms. Another type of algorithms were also developed to form the A-Teams. These A-Teams developed were joined to another one, based in construction heuristics. Some instances of the JSP were tested and the results obtained show the adequability oí this technique to solve this problem.<br>Mestrado<br>Mestre em Ciência da Computação

This thesis investigate a complex real world job shop scheduling / rescheduling problem, in which the presence of uncertainties and the occurrence of disruptions are tackled to produce efficient and reliable solutions. New orders arrive every day in the shop floor and they have to be integrated in the existent schedule. Match-up algorithms are introduced to collect the idle time on machines and accommodate these newly arriving orders. Their aim is to obtain new schedules with good performance which are at the same time highly stable, meaning that they resemble as closely as possible the initial schedule. Subsequently, a novel approach that combines these algorithms with a fuzzy robust scheduling system is proposed. The goal is to associate an effective repairing mechanism with the production of initial robust schedules that are able to facilitate the accommodation of future disruptions. Statistical analyses reveal that match-up algorithms are effective repairing strategies for managing complex disruptions, in which high quality stable schedules are delivered. Moreover, their combination with fuzzy robust scheduling has a positive effect on responding to these disruptions leading to even more reliable solutions in a real world dynamic and uncertain shop floor.