No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Data of assembly line balancing problems
... Various small, medium, and large datasets are available for download, from which three problems are selected for each size category. Table 3 provides detailed information on the benchmark ALB problems (Scholl 1993). The selected data incorporates actual conditions to align with real-life production line scenarios and includes additional data elements such as equipment power ratings. ...
... i. Computational experiment purpose: Scholl (1993). Benchmark data sets by Scholl (1993). ...
... i. Computational experiment purpose: Scholl (1993). Benchmark data sets by Scholl (1993). https:// assem bly-line-balan cing. ...
Assembly line balancing is assigning tasks to workstations in a production line to achieve optimal productivity. In recent years, the importance of energy studies in assembly line balancing has gained significant attention. Most existing publications focused on energy consumption in robotic assembly line balancing. This paper focuses on assembly line balancing with energy consumption in semi-automatic operation. The algorithm serves to improve the exploration to achieve a high-quality solution in a non-convex combinatorial problem, such as assembly line balancing with energy consumption. A novel approach called the Substituted Tiki-Taka Algorithm is introduced by incorporating a substitution mechanism to enhance exploration, thus improving the combinatorial optimization process. To evaluate the effectiveness of the Substituted Tiki-Taka Algorithm, a computational experiment is conducted using assembly line balancing with energy consumption benchmark problems. Additionally, an industrial case study is undertaken to validate the proposed model and algorithm. The results demonstrate that the Substituted Tiki-Taka Algorithm outperforms other existing algorithms in terms of line efficiency and energy consumption reduction. The findings from the case study indicate that implementing the Substituted Tiki-Taka Algorithm significantly increases line efficiency while simultaneously reducing energy consumption. These results highlight the potential of the proposed algorithm to positively impact manufacturing operations by achieving a balance between productivity and energy efficiency in assembly line systems.
... Fuente: Scholl (1993) ...
... Si bien las metodologías que se han propuesto muestran el desarrollo del método para llegar a la solución, no enseñan valores específicos para la media y la varianza de las tareas. En tal sentido, Armin Scholl (1993) propuso un conjunto de problemas, los cuales han sido utilizados por distintos autores en soluciones al problema de balanceo de líneas; sin embargo, resulta complicado encontrar problemas en la literatura que muestren los valores específicos para la varianza de las tareas; en consecuencia, fue necesario desarrollar un método y combinarlo con el enfoque de Carraway utilizado por Urban y Chiang (2006) para la generación de dichas varianzas. Los pasos para la solución del cromosoma generado se describen a continuación: ...
... Para evaluar el algoritmo para el problema de balanceo de líneas en forma de U tipo 1 estocástico se utilizó el conjunto de problemas de balanceo de línea presentado por Armin Scholl (1993), el cual ha sido utilizado por varios autores para probar diferentes metodologías de solución al problema de balanceo de línea. Este conjunto de problemas propone tiempos de tarea, los cuales fueron considerados como la media del tiempo de la tarea (µi). ...
Actualmente, la mayoría de investigaciones acerca del problema de balaceo de líneas de ensamble consideran que los tiempos de las tareas son determinados. Sin embargo, en los procesos de fabricación siempre existe la posibilidad de obtener en los procesos variaciones que impactan en los tiempos de las tareas. Por eso, en el presente trabajo, con base en un enfoque estocástico, se presenta un método que utiliza técnicas metaheurísticas mediante un algoritmo genético, el cual tiene como objetivo brindar una solución al problema de balanceo tipo 1 de líneas en forma de U con tiempos de tarea estocásticos. Para ello, se han tomado como referencia problemas existentes en la literatura para luego ofrecer una comparación entre las soluciones existentes. En el proceso de validación se utilizaron siete categorías de problemas resueltos por otro método. La solución brindada por el algoritmo se sometió a un análisis experimental de los datos para comprobar si era capaz de dar una o más soluciones mejores a las existentes; de ese modo, se buscó balancear la línea con la menor cantidad de recursos humanos posible. Los datos muestran mejores soluciones para los problemas de alta varianza únicamente en el resultado WS mayor, donde se observa una diferencia del 4 %; en los demás hallazgos los porcentajes son mejores. Además, se encontraron seis soluciones mejores a las existentes.
... We evaluated the rebalancing performance of our proposed heuristic with well-known test problems of Scholl (1993Scholl ( , 1999. Our problem set contains 25 test problems in which the number of tasks is between 7 and 297, and cycle times are between 7 and 5853 seconds. ...
... We evaluated the rebalancing performance of our proposed heuristic with well-known test problems of Scholl (1993Scholl ( , 1999. Our problem set contains 25 test problems in which the number of tasks is between 7 and 297, and cycle times are between 7 and 5853 seconds. ...
... benchmark data family of the precedence graph(Scholl, 1993), 2 results of the hybrid genetic algorithm ofBelassiria et al.(2018), m * : the minimal number of stations for the new cycle time, a the proposed method found a better solution than h-GA(Belassiria et al., 2018) ...
This paper recommends a new kind of assembly line rebalancing and worker assignment problem, taking ergonomic risks into account. Assembly line rebalancing problem (ALRBP) occurs when a current line must be rebalanced due to conditions such as changes in demand, production processes, product design, or quality issues. Although there are several research attempts on ALRBP in the relevant literature, only a few studies consider workers as unique individuals. This paper aims to solve the double reassignment problem: tasks to workers and workers to stations, considering ergonomic risk factors. This paper is the first study that comprises worker assignment and ergonomic constraints in ALRBP literature to the best of our knowledge. Objectives of our novel problem are to minimize rebalancing cost, which includes transportation of tasks and workers and minimize stations' ergonomic risk factors. A randomized constructive rule-based heuristic approach is developed to cope with the problem. The proposed solution approach is applied to benchmark data, and obtained results are promising. Moreover, the proposed solution approach is implemented in an automotive parts manufacturing plant.
... For selecting tasks in a workstation, a decision variable f i defines the finish time of job i and relates to the sequencing of tasks in a workstation. This finish time is subject to cycle time constraint in Equation (12) and processing time constraint in Equation (13). ...
... The model has been implemented in the CPLEX optimization package to obtain results using the exact method. Initial ALBP precedence networks are taken from [13] that are also available at the www.assembly-line-balancing.de website. These initial data are further enriched by new parameters of the newly developed model. ...
... If the model is correct, the solution must be the same as simple-ALBP type 1 (SALBP1). In these tests, using some precedence networks taken from [13], the answers exactly match those of SALBP1 counterparts. These conclude that the mathematical model is verified. ...
Human-robot collaboration (HRC) has arisen as a promising technology to improve the productivity of assembly processes. This paper discusses an assembly line balancing problem (ALBP) where manual, robotic, or HRC operations may be considered decision alternatives. Each assembly process task may be operated either by a human operator, a robot operator, or an HRC. This possibility of shared functions between humans and robots may result in a hybrid manual-robotic assembly line. This problem’s mathematical model is developed based on the simple ALBP and modifying the idea of two-sided ALBP, with additional aspects related to resource alternatives of human, robot, or HRC, and robot’s tool-type for the operations. The problem is formulated analytically in a mixed-integer linear programming model with a cost-oriented objective function. The exact method can be applied to obtain an optimal solution.
... When solving instances of SALBP-E, bounds on the number of stations are usually given externally (denoted as m e or m e ), as in Scholl's (1993) dataset. However, extended to TSALBP-1/2, one can state, more generally ...
... Subsequently, the solution methodology that yields the most promising results is used to solve new TSALBP-1/2 instances in Subsection 5.5.2. All tests are performed on the benchmark dataset described by Scholl (1993). This benchmark dataset consists of 24 datasets, each with a different number of instances, where an instance is characterized by the lower and upper bounds on the number of stations. ...
... Divergence (div) and convergence (conv) indicate how strongly/weakly the precedence graphs converge and diverge. For further information and additional metrics, the interested reader is referred to Scholl (1993) and Otto et al. (2013). Finally, the number of instances (#inst) is given in the table. ...
... A second dataset containing larger problems in terms of number of tasks and stations is also proposed. This second dataset is based on the instances from Scholl (1993) , containing 25 precedence graph ranging from 8 to 297 tasks. One instance is generated for each of the 25 precedence graphs. ...
... The 25 instances are generated with P L (average processing load) of 0.95 and an LM (station length multiplier) of 1.5. The number of stations is defined as an average of all instances of Scholl (1993) dataset. If any instance was proven infeasible, new random values for the time distributions were drawn. ...
... A third dataset containing 10 instances is used to explore the effect of the number of demand scenarios and the length of the MPS. For this dataset, the precedence graph of 'Hahn' is used ( Scholl, 1993 ). The generation of the models is identical to the second dataset, but the number of demand scenarios is increased. ...
The quality of the balancing of mixed-model assembly lines is intimately related to the defined production sequence. The two problems are, however, incompatible in time, as balancing takes place when planning the line, while sequencing is an operational problem closely related to market demand fluctuations. In this paper, an exact procedure to solve the integrated balancing and sequencing problem with stochastic demand is presented. The searched balancing solution must be flexible enough to cope with different demand scenarios. A paced assembly line is considered and utility work is used as a recourse for station border violations. A Benders’ decomposition algorithm is developed along with valid inequalities and preprocessing as a solution procedure. Three datasets are proposed and used to test algorithm performance and the value of treating uncertainty in mixed-model assembly lines. The integration of the strategic balancing problem with the operational sequencing problem results in more robust assembly lines.
... In order to assess the performance of our FLDP solution approaches, we evaluate a total of 528 instances. We base our instances on the data set of Scholl (1993). This data set is commonly used as benchmark for simple assembly line balancing problems. ...
... Since the data set of Scholl (1993) is used for the simple assembly line balancing problem, the instances only consist of a single model. We use a procedure similar to that of Li and Gao (2014) to derive multi-model instances. ...
... In our base case (sh = 0%), all models need all tasks. Thus, the structures of the models' precedence graphs are identical and match the precedence graphs in the data set of Scholl (1993). For sh = 25%, we randomly remove 25% of the task-model assignments in the base case. ...
The increasing vehicle heterogeneity is pushing the widespread mixed-model assembly line to its limit. The paced, serial design is incapable of coping with the diversity in workloads and task requirements. As an alternative, the automotive industry has started to introduce flexible layouts for segments of the assembly. In flexible layouts, the stations are no longer arranged serially and no longer linked by a paced transportation system but by automated guided vehicles. This paper investigates the initial configuration of such systems. The flexible layout design problem (FLDP) is the problem of designing a flexible layout for a segment of the assembly of heterogeneous vehicles. It comprises an integrated station formation and station location problem. Moreover, the FLDP anticipates the operational flow allocation of the automated guided vehicles. We formalize the FLDP in a mixed-integer linear program and develop a decomposition-based solution approach that can optimally solve small- to mid-sized instances. In addition, we transform this solution approach to a matheuristic that generates high-quality solutions in acceptable time for large-sized instances. We compare the efficiency of flexible layouts to mixed-model assembly lines and quantify the benefits of flexible layouts which increase with vehicle heterogeneity.
... Difficulty levels of test instances vary according to their characteristics. These are; number of tasks, task times, cycle times, precedence relationships, number of stations and the attributes of task times (Scholl, 1995). Depending on these characteristics, in the literature, various indicators have been developed to measure the difficulty of a given balancing problem instance. ...
... The time variability ratio is obtained by dividing the maximum task duration by the minimum (t max /t min ). A low time variability ratio indicates that the tasks are close to each other, while a high ratio indicates that the variability between durations is high (Scholl, 1995). In this section two different computational studies were carried out. ...
... Bu nedenle, çalışmada yalnızca Mitchell örneklemine ait öncelik ilişkisi diyagramına yer verilmiştir. Mitchell örneklemi için işlem süreleri ve görevler arasındaki öncelik ilişkilerini gösteren ağ yapısı [46] Şekil 7'de verilmiştir. Bu ağ yapısında düğümler görevleri, bağlantılar da görevler arasındaki öncelik ilişkilerini göstermektedir. ...
... Bu ağ yapısında düğümler görevleri, bağlantılar da görevler arasındaki öncelik ilişkilerini göstermektedir. 13(5) 12(1) 10(1) 11(3) 16(3) 15 (5) 17 (13) 20 (3) 7 (8) 14 (3) 18 (5) 19 (2) 1(4) 2(3) 3(9) 4(5) 21 (7) 5(9) 6(4) 8(7) 9(5) Şekil 7. Mitchell örneklemi için işlem süreleri ve öncelik ilişkisi diyagramı (Precedence and process times diagram for Mitchell data set) [46] 5. ...
Montaj hatlarında yeniden işleme istasyonları genellikle uygun olmayan ürünlerin yeniden işlendiği bir istasyon olarak kullanılmaktadır. Bununla birlikte, yeniden işleme istasyonunun yalnızca bu amaç için kullanılması montaj hattı hata oranının düşük olması durumunda, bu istasyonda bulunan kaynakların verimsiz bir şekilde kullanılmasına neden olabilmektedir. Bu çalışmada, öncelikle, yeniden işleme istasyonunun paralel görevler için kullanılması dikkate alınarak çevrim süresini minimize etmeyi amaçlayan bir doğrusal olmayan karışık-tamsayılı programlama modeli sunulmuştur. Paralel görevlerle ilgili doğrusal-olmayan kısıt, değişken dönüşümü ile doğrusallaştırılarak, model doğrusal-karışık-tamsayılı şekle dönüştürülmüştür. Buna ek olarak, çalışma kapsamında, problemin stokastik unsurları dikkate alınarak, tamsayılı programdan elde edilen sonuçların doğrulanması için de bir simülasyon modeli geliştirilmiştir. Geliştirilen model ile örnek bir problem simüle edilerek, söz konusu problem için tamsayılı programlama modelinden elde edilen çözümün uygulanabilirliği incelenmiştir.
... They aimed to minimise M while minimising SI and maximising E. Based on the concepts of adaptive learning and SA, this paper proposes a hybrid improvement heuristic solution for SLs and USLs. A series of comparative tests are performed on the benchmark problems presented by Scholl (1995). With the exception of one difficult case, the proposed approach identified the best solution. ...
Purpose: Assembly Line Balancing (ALB) is critical to manufacturing efficiency and productivity. It involves assigning tasks to workstations to optimise performance while satisfying task priority and cycle time constraints. The Simple ALBP (SALBP) is a simplified version of the general problem that has received considerable research interest. Many academic works have been published on this topic, using a variety of methods, including exact, heuristic, and metaheuristic approaches. Therefore, the purpose of this research is to present a comprehensive evaluation of the literature on the methods used to solve the SALBP.Design/methodology/approach: A comprehensive literature review was conducted to identify, select, analyse, and summarise 126 papers on SALBPs. The study started with the selection of relevant keywords. The selected papers were then narrowed down using various criteria.Findings: The analysis showed that SALBP-1 and SALBP-2 are the most common types, with metaheuristic approaches being the most widely used. Despite extensive research, there is a significant need for studies focusing on SALBPs for multi- and mixed-models, particularly in the context of U-shaped and two-sided lines.Originality/value: This literature review contributes to the identification of key areas for improvement in the SALBP and provides insight into potential directions for future research.
... This subsection provides a comprehensive overview of the instances generated, covering both the SALBPGen procedure proposed by Otto et al. (2013) for SALBP and the specific generation process for PALBP-B instances. Otto et al. (2013) highlight that the classic instances of the SALBP literature, such as those collected in Scholl (1995) and Scholl (1999), suffer from several deficiencies, such as being poorly diversified and randomised. The authors proposed the SALBPGen procedure to generate instances according to specific guidelines and principles such as the structure of precedence graphs, order strength, and task time distribution. ...
... Time Variability (TV) since SALPB 1 relies on tasks rather than assigning workers to workstations, TV is a critical factor in ALB. Scholl in 1995 [57] and Huo et al. in 2018 [58] measure TV as the ratio of the longest task duration to the shortest task duration. Nonetheless, this ratio does not represent the variation of all tasks. ...
A great deal of work has been directed to assembly line balancing (ALB) using different heuristics. However, little work is made on evaluating the performance of ALB heuristics under various operating factors. In this paper, the performance of common heuristics in solving the single-model assembly line balancing problem (SLABP-1) is investigated using a dataset of randomly generated ALB problems. The dataset is sorted to obtain representative levels of cycle time, time variability (TV), number of tasks, and order strength. An experiment is then designed to investigate the effects of factors on the performance of heuristics with a focus on line efficiency, largest workstation idle time, smoothness index, total idle time, number of workstations, and west ratio. In this study, the optimal working settings for the balancing process are determined based on the performance metrics achieved. The results show that several heuristics can optimize multiple performance metrics given specific factor settings. Furthermore, findings reveal that some performance measures can be optimized at the same factor levels regardless of the heuristics used. TV is recognized to be a critical factor to ALB problem, with lower TV achieving better performance measures. Process designers should understand the impact of factors effects on choosing the ALB heuristic. A heuristic, which can optimize multiple performance measures simultaneously, is recommend.
... The presented algorithm is evaluated on the basis of practical problems from car body construction. The BARTHOLD problem statement from the data sets by Scholl [11] is chosen because it matches the F-Ratio (0.7 to 0.8) and WEST-Ratio (15 to 25) (see Dar-El [12]), which we have found to be typical for car body construction problems. The cycle times and processing times of the problem are multiplied by 0.1 s, which results in cycle times of 56.4 s, 62.6 s, 70.5 s, and 80.5 s. ...
In this paper, a genetic algorithm for the robotic assembly line balancing problem (RALBP) is developed that supports multimodal stochastic processing times and multiple parallel-working robots per workstation. It has the objective to minimize the amount of workstations at a given production rate and probability limit for violating the cycle time (PL). The algorithm is evaluated on the BARTHOLD data set in a range of 1 % to 50 % for PL using an experimentally determined and a normal distribution for the task times. The increase of PL results in a shift of tasks from rear to front stations, because more tasks can be assigned to each station. The shift using normal distributed task times is stronger. This demonstrates the importance of realistic stochastic distribution assumptions. For practical applicability, more constraint types have to be included in the future.
... To be able to make a fair comparison, we report CPU times instead of the clock times and set the CPU time limit to 3600 s for all experiments. Five well-known problem instances from the assembly line balancing literature are used: 25-task problem from the study of Nkasu and Leung (1995); Sawyer's 30-task, Gunther et al.'s 35-task, Kilbridge and Wester's 45-task, and Tonge's 70-task problems (Scholl 1995). Precedence diagrams of the problem instances are used directly, and deterministic task processing times in the benchmark problems are used as the mean task processing times ( i2 ) for human workers. ...
To keep up with the Industry 4.0 technological revolution and get the upper hand over competitors, manufacturing companies replace human workers with robots in their assembly processes. A popular approach in the manufacturing industry is to design an assembly line with human-robot collaboration. In this study, we investigate a robotic stochastic assembly line balancing problem (RSALBP), with the motivation to observe the effects of robots on the cycle time in stochastic assembly lines where human workers and robots operate in different workstations. In the literature, robotic assembly line balancing is only studied with deterministic task times. However, assembly line balancing contains stochastic processes in real life. We assume that the processing time of each task follows a normal distribution whose parameters depend on the type of the operator performing the task with robots having much less (possibly zero) variation in task times than human workers. It is assumed that human workers are fully capable while robots are able to perform a subset of the tasks. We study type-II RSALBP which aims to minimize the cycle time for an assembly line with stochastic task times, given a fixed number of workstations and robots. This problem is NP-hard and includes non-linearity. We propose a mixed-integer second-order cone programming formulation and a constraint programming formulation to solve the problem. Instances from the literature are used to test the effectiveness of the proposed formulations. Additionally, the effects of robots on cycle times are evaluated by conducting a computational study with a comprehensive experimental design.
... Among those, SALOME is an effective bidirectional branch-and-bound (B&B) procedure developed by Scholl and Klein (1997) to solve the SALBP. It can solve up to medium-size problems to optimality, but still struggles with some large instances in a well-known benchmark dataset (Scholl, 1993). More recently, Sewell and Jacobson (2012) proposed a branch, bound, and remember (BBR) algorithm and Morrison et al. (2014) subsequently added a new backtracking rule to save memory in their own BBR variant. ...
The assembly line balancing problem (ALBP) assigns tasks to (work)stations to manufacture products. It divides the station loads as evenly as possible since any bottleneck defines the production rate of a flow shop system. The sum of task processing times represents the station load in the classical Simple Assembly Line Balancing Problem (SALBP). Thus, the most loaded station imposes a lower bound on the line's cycle time. However, the simple sum of processing times is only valid under several assumptions. For instance, more realistic ALBPs may contain stochastic data, multiple workers per station, or depend on the production sequence. Hence, the station load computation involves further scheduling decisions for this latter class of problems. In order to tackle these ALBPs with practical extensions, several literature contributions have recently employed Benders decomposition (BD) algorithms. The BD structure allows a division of the formulation into two or more levels. More specifically, this framework permits the decoupling of task assignment decisions and cycle time assessments. In the field of ALBPs, various stochastic and sequence-dependent problems have successfully applied this approach. This paper provides a literature review of these methods, the different formulations, implementation details, and improvement ideas.
... Plus les liens de précédences sont nombreux pour une quantité donnée de tâches N , moins il y a de degrés de liberté possible pour créer des solutions, [105], et potentiellement elles deviennent plus dureà identifier par les algorithmes [106]. On utilisera comme indicateur 'l'order strength' OS, qui est défini par le ratio du nombre de relations de précédences présentes Q par le nombre maximal de relations possibles P . ...
La robotique collaborative est une technologie disponible pour l'industrie, et ouvre de nouvelles perspectives de création de process via de nouvelles possibilités d’interaction homme-machine. Or les résultats d'implémentation peuvent apparaitre contrastés en terme de performance, ce qui pose le problème d’efficacité de l’implémentation de technologies innovantes dans les process de production. A contrario, le lean manufacturing, sans être une approche techno-centrée, est aujourd’hui reconnu comme une méthode efficace quand il s’agit de dégager une performance durable des process de production, notamment en mettant à contribution les acteurs du process dans des activités d’amélioration continue. L’objectif de ce travail est d’évaluer la pertinence de ces techniques issues du lean pour l’intégration de robotique collaborative, dans le contexte même de leur mise en œuvre, c’est-à-dire en s’appuyant sur les compétences développées par les opérateurs. Pour cela ce travail de recherche s’appuie sur les méthodes de recherche orientée conception, utilisées notamment en science de l’éducation. Nous avons donc développé et testé avec des groupes d’étudiants un protocole expérimental qui simule l’intégration de robot collaboratif dans un process d’assemblage initialement manuel. Cette démarche pourra ensuite être exploitée comme stratégie d’implémentation de robot collaboratif. Différents cas d’études de complexité variable sont testés. De plus, un algorithme d’optimisation a été spécifiquement développé afin de possiblement prendre le relai des techniques lean qui n’ont pas originellement été conçues pour intégrer des moyens aussi reconfigurables que les robots collaboratifs. Une évaluation de cet outil de résolution est réalisée, ainsi que des propositions d’évolution afin d’enrichir la méthode proposée.
... The mathematical programming models presented in this paper are not easy to solve, especially in scenarios that are on a practical scale. To evaluate the computational performance of the presented models, 11 datasets from Scholl (1993) were used; each dataset is referred to by its name. Tasks in these problems are denoted by integer numbers. ...
Adding robots to a human-operated assembly line influences both the short- and long-term operation of the line. However, the effects of robots on assembly line capacity and on cycle time can only be studied if appropriate task assignment models are available. This paper shows how traditional assembly line balancing models can be changed in order to determine the optimal number of workstations and cycle time when robots with different technological capabilities are able to perform a predetermined set of tasks. The mathematical programming models for the following three cases are presented and analysed: i) only workers are assigned to the workstations; ii) either a worker or a robot is assigned to a workstation; iii) a robot and a worker are also assigned to specific workstations. The data of an assembly line producing power inverters is used to illustrate the proposed calculations. Both the assignment of tasks and the changes of cycle time are analysed within the AIMMS modelling environment. The computational characteristics of the proposed mathematical programming models are also examined and tested using benchmark problems. The models presented in this paper can assist operations management in making decisions relating to assembly line configuration.
... Our new DE algorithm was evaluated using data retrieved from Scholl [33], which included 15 problem sets (101 instances): 10 medium problem sets with 21-58 tasks (50 instances) and 5 large problem sets with 70-297 tasks (51 instances). The UALBP-2 was solved by applying the java program in NetBeans software environment on a computer with an Intel i5-8500 CPU@3.0 ...
... 128 test instances which are proposed by Scholl (1993) and which are available at https ://assem bly-line-balan cing.de/ have been tested and the computational results have been compared with Blum (2011). ...
This paper presents the novel method variable neighbourhood strategy adaptive search (VaNSAS) for solving the special case of assembly line balancing problems type 2 (SALBP-2S), which considers a limitation of a multi-skill worker. The objective is to minimize the cycle time while considering the limited number of types of machine in a particular workstation. VaNSAS is composed of two steps, as follows: (1) generating a set of tracks and (2) performing the track touring process (TTP). During TTP the tracks select and use a black box with neighborhood strategy in order to improve the solution obtained from step (1). Three modified neighborhood strategies are designed to be used as the black boxes: (1) modified differential evolution algorithm (MDE), (2) large neighborhood search (LNS) and (3) shortest processing time-swap (SPT-SWAP). The proposed method has been tested with two datasets which are (1) 128 standard test instances of SALBP-2 and (2) 21 random datasets of SALBP-2S. The computational result of the first dataset show that VaNSAS outperforms the best known method (iterative beam search (IBS)) and all other standard methods. VaNSAS can find 98.4% optimal solution out of all test instances while IBS can find 95.3% optimal solution. MDE, LNS and SPT-SWAP can find optimal solutions at 85.9%, 83.6% and 82.8% respectively. In the second group of test instances, we found that VaNSAS can find 100% of the minimum solution among all methods while MDE, LNS and SPT-SWAP can find 76.19%, 61.90% and 52.38% of the minimum solution.
... In this part, our experiments use three referenced data sets of SALBP, which is data set of Talbot et al. [39], Hoffmann [40] and Scholl [41] respectively. The reported best-known solutions of the instances studied in these data sets provide a benchmark for numerical computations. ...
Considering the characteristics of multimodels production pattern on assembly line, the assembly line balancing problem which is non-deterministic polynomial hard becomes more challenging to complete. In this article, we propose a reformulation of simple assembly line balancing problem based on Dantzig-Wolfe decomposition. New models of the master problem and subproblems in this algorithm are built.We implement a branching rule which is suited to seeking integer solutions of the problem. A new three-stage branch-and-price algorithm is designed to accelerate the process of searching the branch-andbound tree. Extensive computational experiments on benchmark data sets, as well as a real industry case are conducted. The numerical results validate the feasibility and effectiveness of the proposed method which performs efficiently on various cases. Effects on optimization results considering the characteristics of the instance sets are analyzed. Results show that the three-stage branch-and-price algorithm is superior to the classic branch-and-price algorithm in terms of solution quality and computing time.
... Data set includes seven problems with task numbers differing between 29 and 111: Buxey (29,8,7,14), Sawyer (30,8,7,14), Gunther (35,10,6,15), Kilbridge (45,9,3,11), Tonge (70, 23,3,25), Arcus1 (83,20,3,22), Arcus2 (111, 25,3,27). The numbers in the parenthesis indicate the task numbers, the number of test instances contained in the relevant ALBP, minimum and maximum number of workstations of test instances in the relevant ALBP, respectively. ...
This study focusses on a two objective type-2 simple assembly line balancing problem. Its primary objective is minimizing the cycle time, or equivalently, maximizing the production rate of the line. Minimization of the workload imbalance among workstations is considered as the secondary objective. Since the problem is known to be intractable, a reactive tabu search algorithm is proposed for the solution. Although tabu search is a well-known meta-heuristic search procedure, based on the detailed literature survey, there is not a reactive tabu search algorithm to solve the investigated problem. Furthermore, the algorithm utilizes a sequence oriented solution representation which is usually applied by population heuristics such as genetic algorithms and differential evolution algorithms. The performance of the algorithm is tested on several benchmark problems taken from the open literature by comparing both objective values with those of previously developed four particle swarm optimization (PSO) algorithms and two multi-objective genetic algorithms (MOGA). The computational results show that the proposed approach presents a quite encouraging success over the existing meta-heuristics.
... The MILP systematic and the heuristics are analysed using benchmark problems given in Scholl and Klein (1999). Details on the test problem can be accessed via Scholl (1993). Test data are categorized as small (# of tasks 30), medium (30<# of tasks 80) and large (# of tasks > 80) with respect to problem size. ...
Purpose
Rabbit chase (RC) is used as one of the most effective techniques in manufacturing systems, as such systems have high level of adaptability and increased productivity in addition to providing uniform workload balancing and skill improving environment. In assembly systems, RC inspires the development of walking worker assembly line (WWAL). On the other hand, U-type assembly lines (UALs) may provide higher worker utilization, lower space requirement and more convenient internal logistics when compared to straight assembly lines. In this context, this study aims to improve assembly line performance by generating RC cycles on WWAL with respect to task assignment characteristics of UAL within reasonable walking distance and space requirement. Therefore, a novel line configuration, namely, segmented rabbit chase-oriented U-type assembly line (SRCUAL), emerges.
Design/methodology/approach
The mathematical programming approach treats SRCUAL balancing problem in a hierarchical manner to decrease computational burden. Firstly, segments are generated via the first linear programming model in the solution approach for balancing SRCUALs to minimize total number of workers. Then, stations are determined within each segment for forward and backward sections separately using two different pre-emptive goal programming models. Moreover, three heuristics are developed to provide solution quality with computational efficiency.
Findings
The proposed mathematical programming approach is applied to the light-emitting diode (LED) luminaire assembly section of a manufacturing company. The adaptation of SRCUAL decreased the number of workers by 15.4% and the space requirement by 17.7% for LED luminaire assembly system when compared to UAL. Moreover, satisfactory results for the proposed heuristics were obtained in terms of deviation from lower bound, especially for SRCUAL heuristics I and II. Moreover, the results indicate that the integration of RC not only decreased the number of workers in 40.28% (29 instances) of test problems in U-lines, but also yielded less number of buffer points (48.48%) with lower workload deviation (75%) among workers in terms of coefficient of variation.
Practical implications
This study provides convenience for capacity management (assessing capacity and adjusting capacity by changing the number of workers) for industrial SRCUAL applications. Meanwhile, SRCUAL applications give the opportunity to increase the capacity for a product or transfer the saved capacity to the assembly of other products. As it is possible to provide one-piece flow with equal workloads via walking workers, SRCUAL has the potential for quick realization of defects and better lead time performance.
Originality/value
To the best of the authors’ knowledge, forward–backward task assignments in U-type lines have not been adapted to WWALs. Moreover, as workers travel overall the line in WWALs, walking time increases drastically. Addressing this research gap and limitation, the main innovative aspect of this study can be considered as the proposal of a new line design (i.e. SRCUAL) which is sourced from the hybridization of UALs and WWAL as well as the segmentation of the line with RC cycles. The superiority of SRCUAL over WWAL and UAL was also discussed. Moreover, operating systematic for SRCUAL was devised. As for methodical aspect, this study is the first attempt to solve the balancing problem for SRCUAL design.
... The data sets of well-known robotic assembly line balancing problems are taken from the literature (Nilakantan et al., 2015;Scholl, 1995) and developed to suit the considered research. As the problem under study is to balance and schedule the material delivery for a robotic assembly line, the problem instances will have the following characteristics: ...
Purpose- Optimizing material handling within the factory is one of the key problems of modern assembly line systems. This paper focuses on simultaneously balancing a robotic assembly line and scheduling of material handling required for the operation of such a system, a topic that has received limited attention in academia. Manufacturing industries focuses on full autonomy due to the rapid advancements in different elements of Industry 4.0 such as internet of things, big data and cloud computing. In smart assembly systems, this autonomy aims at the integration of automated material handling equipment like automated guided vehicles (AGVs) to robotic assembly line systems to ensure a reliable and flexible production system.
Design/methodology/approach- This paper tackles the problem of designing a balanced robotic assembly lines and scheduling of automated guided vehicles to feed materials to these lines such that the cycle time and total tardiness of the assembly system are minimized. Due to the combination of two well-known complex problems: line balancing and material handling, a heuristic and metaheuristic-based integrated decision approach is proposed.
Findings- Detailed computational study demonstrates how integrated decision approach can serve as an efficient managerial tool in designing/redesigning assembly line systems and support automated transportation infrastructure.
Originality/value- This work is beneficial for the production managers in understanding the main decisional steps involved in the designing/redesigning of smart assembly systems and providing guidelines in decision making. Moreover, this study explores the material distribution scheduling problems in assembly systems, which is not yet largely explored in the literature.
... In the literature, there are no similar studies considering minimizing both cycle time and number of resources required. Hence, we select sample assembly line model from Scholl (1993) and modify the problems by assigning type of resource(s) required for each task as in scenario (ii) and (iii). Five problem instances consisting of 21-70 tasks from literature are selected. ...
The literature studies assume that resources used to be perform the tasks are certain and homogenous in any assembly line. However, tasks may need to be processed by general resource requirements in real life. These general resources could be classified by usage of resources such as simple or multiple, alternative and concurrent. The problem which is related to assignment of the task to any workstation and assignment of resources needed by the task simultaneously is defined as resource constrained assembly line balancing problems (RCALBP). In this study, a multi-objective model with minimization of cycle time and resource usage for a given number of stations is modelled to solve the RCALBP for the first time. Alternative and general resource types for tasks and using more than two resource type requirements are also considered. A constraint programming model is developed and solved to find the optimal solutions of these problems. The proposed models are tested with sample scenarios to show the effectiveness of the model. We would like to thank the referee for taking the time to review our article. According to review the arrangements are given as follows in bold type:-the error in CP model. The error in CP model is corrected. Equation 12 is rewritten.-computational times or complexity Regarding the solution time, the following text is added to Section 5. "Even in the solution of large problem sets, CPU time is measured in seconds (eg. Tongue 70 tasks 5 resources problem CPU time is 1.50 s). For this reason, it was not reported in the results of numerical experiments." Regarding the complexity, the following text is added to Section 2. "As mentioned in Agpak and Gokcen (2005), Corominas et al. (2011) and Quyen et. al. (2017), the RCALBP is an NP-Hard problem due to its combinatory nature." ABSTRACT The literature studies assume that resources used to be perform the tasks are certain and homogenous in any assembly line. However, tasks may need to be processed by general resource requirements in real life. These general resources could be classified by usage of resources such as simple or multiple, alternative and concurrent. The problem which is related to assignment of the task to any workstation and assignment of resources needed by the task simultaneously is defined as resource constrained assembly line balancing problems (RCALBP). In this study, a multi-objective model with minimization of cycle time and resource usage for a given number of stations is modelled to solve the RCALBP for the first time. Alternative and general resource types for tasks and using more than two resource type requirements are also considered. A constraint programming model is developed and solved to find the optimal solutions of these problems. The proposed models are tested with sample scenarios to show the effectiveness of the model.
... In this subsection, we present ten test problems. The precedence diagram of the tasks with the tasks time for each of these test problems is taken from [48]. Some data of these test problems are presented in Table 1. ...
In this study, a new cost-oriented assembly line balancing problem is proposed and formulated. A single objective function consisting of minimizing the cost associated with equipment, labor wage, and station establishment is considered for the problem. This problem is more complicated comparing to the literature as worker qualification is considered for determining his/her wage. As this problem is of NP-hard optimization problems, some meta-heuristic solution approaches, e.g., simulated annealing, variable neighborhood search, genetic algorithm, tabu search, population-based simulated annealing, and their hybrid versions are proposed. In the proposed algorithms, a novel encoding–decoding scheme is applied. This scheme uses the Hungarian method to assign the workers to the station to reduce the total wage of the workers. To study the performance of the proposed meta-heuristic algorithms, ten test problems are generated randomly, and using one of them the parameters of the algorithms are tuned by the Taguchi method. The final experiments on the proposed algorithms and the test problems show that in the most of the experiments, among the proposed algorithms, the single-solution-based algorithms, except TS, perform better than the population-based algorithms, especially for the case of large size test problems.
Efficient configurations of assembly lines involve balancing the workload across stations and strongly depend on the sequence in which the products are processed. While these processes are interconnected, they are typically addressed separately due to the complexities and different time frames involved. Even if some literature methods combine balancing and sequencing problems, these are limited in scalability, often constrained to small-scale instances with a limited number of products. This restricts their applicability to real-world industrial scenarios, where the number of product configurations can be vast. Another branch of the literature assumes a random sequence, which is a pessimistic view of a real setting. To fill this gap, we propose a genetic algorithm for the integration of the balancing problem with a semi-random production sequence that respects selected car-sequencing rules. This paper addresses the combined balancing and car sequencing rules selection problem for large-scale assembly lines with uncertain production sequences. By incorporating rules related to car sequencing, we reduce the randomness of a simulated production sequence used to evaluate the efficiency of an assembly balancing solution without the burden of solving an optimization problem. The impact of incorporating rules is evaluated through a discrete-event simulator, demonstrating significant improvements in line performance.
Son yıllarda, tam zamanında üretim felsefesini benimseyen işletmelerin artmasıyla birlikte U-şekilli montaj hatlarının kullanımı yaygınlaşmıştır. Dolayısıyla, U-şekilli montaj hattı dengeleme problemlerinin çözüm yöntemleri üzerine yapılan çalışmalar çoğalmaktadır. Bu çalışmada, iş yükü düzgünleştirme amacını ikincil amaç olarak dikkate alan, tip-2 U-şekilli basit montaj hattı dengeleme probleminin çözümü için, literatürde daha önce geliştirilmiş ve farklı problemler üzerinde etkinliği gösterilmiş olan, iki adet farklı yapıda melez tavlama benzetimi-tabu arama algoritması kullanılmıştır. Biri tavlama benzetimi, diğeri tabu arama üzerine inşa edilmiş bu algoritmalar ilk defa U-şekilli bir montaj hattı dengeleme problemine uygulanmaktadır. Bunun yanında, bu iki melez algoritma, performans karşılaştırma amacıyla, ilk defa bir problemin çözümünde beraberce kullanılmıştır. Melez algoritmaların performansları, literatürden alınmış test problemleri üzerinde, birbirleriyle ve tavlama benzetimi ve tabu arama algoritmalarının saf versiyonlarıyla karşılaştırılmıştır. Hesaplama sonuçları, çekirdeğini tabu aramanın oluşturduğu melez algoritmanın diğerlerine üstünlük sağladığını göstermektedir.
The semiautomated assembly line balancing problem (SAALBP) is a complex planning problem in which the production process comprises human workers, robots, and human-robot collaboration. With increasing awareness of sustainable practices in the manufacturing environment, this study addresses a sustainable SSALBP, where a mixed integer linear programming (MILP) model is formulated for the first time for this type of assembly systems. The model consists of three objectives and each of the objectives represents the three pillars of sustainability. The problem becomes more realistic by considering uncertainty in processing time and the influence of human workers’ age and skill in processing time. The uncertainty is adopted in the MILP model by using a chance-constraint programing approach. Since the problem is complex, Q-learning and Monte-Carlo simulation-assisted based genetic algorithm (GA) based memetic algorithm (MA) are proposed to solve the problem. The performance of MA is evaluated against the well-known non-dominated sorting GA-III (NSGA-III) and state-of-the-art algorithms for solving the problem. The experimental results show that MA surpasses its contest significantly in finding better Pareto-fronts.
Consumer behavior has undergone a major transformation in recent years. On-
demand has become one of the most important expectations. There is a very short time available in passenger and freight areas to process, order and meet incoming needs. In order for service providers to meet this through the efficient use of available resources, they need to take a new planning approach to maintain their efficiency and maintain a high standard of service. Within passenger transport, there is a sub-sector that has received less emphasis in recent decades. And this is patient transport. Sick people have the opportunity to use patient transport services, but the route organization of these services is not efficient enough, the service is expensive. Yet there would be a need for a service that can help people travel to hospitals at the right level. In this article, we present the solutions available
abroad and a suitable mathematical model that can solve the transport organization task quickly and efficiently in the case of a vehicle. We also take into account the fact that patients can set their own maximum travel time and it is not up to the operating company to decide. We also pay attention for that there are patient who needs more space on the vehicle (e.g. wheelchair users), so the passengers are heterogeneous in this system. The usability of the model is also illustrated through an example.
To tackle the balancing and sequencing problems of flexible mixed model assembly lines with alternative precedence relations, If–then rules and AND/OR graphs are adopted as modelling tools to replace precedence graphs that have limitations in representing alternative precedence relations. Mixed integer linear programming (MILP) and constraint programming (CP) models are established respectively. Moreover, an iterative decomposition method is developed to deal with large-sized problems. Computational experiments on different scales are carried out to test those methods. The computational results reveal that the performance of CP is better than MILP especially when more OR relations exist. And the iterative decomposition method outperforms MILP and CP in terms of solution quality and time.
This paper emphasises the parallel adjacent U-shaped assembly line balancing problem in which the collaborative work forces in the system comprise human workers and robots. Apart from normal workers, disabled workers are also hired to enable the company to enjoy corporate tax benefits from the government. Five objectives related to the system as a whole and another three interacting entities are optimised simultaneously. A mathematical model is formulated for the first time for this type of layout. Due to the NP-hard nature of the problem, a reference point based evolutionary algorithm, namely non-dominated sorting teaching-learning-based optimisation (NSTLBO III), is proposed to solve medium- and large-sized problems. The performances of NSTLBO III are benchmarked in the sense of Pareto efficiency with two well-known multi-objective evolutionary algorithms, i.e. non-dominated sorting genetic algorithm (NSGA III) and multi-objective evolutionary algorithm based on decomposition (MOEA/D). The experimental results show that NSTLBO III surpasses its contestants significantly concerning the convergence to the true Pareto front.
We analyze the assembly line balancing problem with stochastic task times. Tasks have to be assigned to a minimum number of stations with a constraint on the line reliability, which is the probability of finishing a work piece completely. A sampling approach is developed that ensures the line reliability. We prove that any lower bound on the number of stations for the related deterministic problem can be transformed into a lower bound for this sampling formulation. This general transformation can be applied to any bound that has already been developed or to any potential new bound. Those bounds can be applied to any MIP model, optimization algorithm or heuristic procedure based on a sampling formulation. We exemplify the usefulness of these bounds in a reliability-based branch-and-bound (RB&B) algorithm that explicitly considers the dependence among all stations due to the constrained line reliability. A partial assignment of tasks to stations has to consider already constructed stations and potential further assignments to other stations. Hence, a feasible assignment of tasks to this station may allow for exceeding the cycle time with a certain probability but has to consider the overall line reliability with respect to the remaining stations. Effective fathoming strategies based on the new transformed lower bounds or based on a direct consideration of the line reliability are proposed. A numerical study shows that the transformed lower bounds are tight and that they substantially reduce the required computation times of the RB&B algorithm and of the solver CPLEX.
Ever since the times of Henry Ford up to today’s industry 4.0 era, flow-oriented assembly processes, where an assembly line conveys the workpieces from workstation to workstation, are very important for mass-producers in manifold branches of industry. Among the most elementary optimization problems in this context is the assembly line balancing problem, which decides on the division of labor among the stations of an assembly line. This paper surveys the scientific literature on assembly line balancing that has been published since the last major review papers have appeared in 2006 and 2007, respectively. We cover all essential stages of the decision making process: we address novel methods to efficiently gather the relevant (precedence graph) data, review especially new problem variants and models treated in the literature, and survey the most important algorithmic developments. Furthermore, we outline a possible research agenda for the next fifteen years.
We study the evolution of rules that define how to assign tasks to workstations in heuristic procedures for assembly line balancing. In assembly line balancing, a set of partially ordered tasks has to be assigned to workstations. The variant we consider, known as the assembly line worker assignment and balancing problem (ALWABP), has a fixed number of machines and workers, and different workers need different times to execute the tasks. A solution is an assignment of tasks and workers to workstations satisfying the partial order of the tasks, and the problem is to find a solution that maximizes the production rate of the assembly line. These problems are often solved by station-based assignment procedures, which use heuristic rules to select the tasks to be assigned to stations. There are many selection rules available in the literature. We show how efficient rules can be evolved, and demonstrate that rules evolved for simple assignment procedures are also effective in stochastic heuristic procedures using beam search, leading to improved heuristics.
We consider a version of the well-known simple assembly line balancing problem (called SALBP-SX) where, given the cycle time and the number of stations, the workloads of the stations are to be leveled according to an adequately defined smoothness index SX. Our index SX involves for each station the quadratic deviation of its workload from the average (or ideal) workload and is therefore closely related to the variance, which is a common measure of dispersion in statistics. Contrary to the existing literature on workload smoothing in ALB, which often treats the optimization of a prespecified smoothness index as a secondary objective, we consider our SX-objective as the single one in order to account for the practical relevance of fair workload distributions and avoiding overloaded bottleneck stations.
To optimally solve SALBP-SX, we develop a tailored branch-and-bound procedure. It contains a new station-oriented branching scheme, new lower bound arguments, logical tests and, in particular, a dynamic programming scheme for the pre-calculation of potential workloads, which accelerates the procedure greatly. In comprehensive computational experiments, we show that our method clearly outperforms a class of recently developed task-oriented branch-and-bound procedures and also the mathematical programming solver Gurobi.
In the literature, most of the researchers studying assembly line balancing have only considered task assignments. However, resources are needed to perform the tasks. Therefore, assigning resources related to tasks becomes more realistic when assigning tasks to stations. In the general case of the problem, the task is performed with a specified amount of resources. If resource types such as a, b, c are required to perform tasks in an assembly line, the combination of tasks required from these resources should also be assigned to the stations. This type of problem is defined as general resources-constrained assembly line balancing problem (GRCALBP). In this study, GRCALBP is addressed to minimize cycle time and resource usage for a given number of stations. New constraint programming (CP) models based on conjunction normal form are proposed. The CP models are tested with generated problem instances from the data set in the literature. The experimental results show that CP is an efficient and effective modeling technique to solve GRCALBP. Finally, suggestions are made regarding alternative objective functions.
In this paper, we investigate the use of the scalability property of RMS to reduce the energy cost during the production. The corresponding optimization problem is a new Bilevel Optimization problem which combines a line balancing problem with a planning problem. A heuristic based on a simulated annealing algorithm and a linear program is proposed. An illustrative example is presented to highlight the potential of this new approach compared to the cost obtained with a classic production line.
The two-volume set IFIP AICT 591 and 592 constitutes the refereed proceedings of the International IFIP WG 5.7 Conference on Advances in Production Management Systems, APMS 2020, held in Novi Sad, Serbia, in August/September 2020.
The 164 papers presented were carefully reviewed and selected from 199 submissions. They discuss globally pressing issues in smart manufacturing, operations management, supply chain management, and Industry 4.0. The papers are organized in the following topical sections:
Part I: advanced modelling, simulation and data analytics in production and supply networks; advanced, digital and smart manufacturing; digital and virtual quality management systems; cloud-manufacturing; cyber-physical production systems and digital twins; IIOT interoperability; supply chain planning and optimization; digital and smart supply chain management; intelligent logistics networks management; artificial intelligence and blockchain technologies in logistics and DSN; novel production planning and control approaches; machine learning and artificial intelligence; connected, smart factories of the future; manufacturing systems engineering: agile, flexible, reconfigurable; digital assistance systems: augmented reality and virtual reality; circular products design and engineering; circular, green, sustainable manufacturing; environmental and social lifecycle assessments; socio-cultural aspects in production systems; data-driven manufacturing and services operations management; product-service systems in DSN; and collaborative design and engineering
Part II: the Operator 4.0: new physical and cognitive evolutionary paths; digital transformation approaches in production management; digital transformation for more sustainable supply chains; data-driven applications in smart manufacturing and logistics systems; data-driven services: characteristics, trends and applications; the future of lean thinking and practice; digital lean manufacturing and its emerging practices; new reconfigurable, flexible or agile production systems in the era of industry 4.0; operations management in engineer-to-order manufacturing; production management in food supply chains; gastronomic service system design; product and asset life cycle management in the circular economy; and production ramp-up strategies for product
This article examines the problem of balancing modern assembly lines for large-size products typical for the automotive industry. The model of a multimanned assembly line balancing problem is proposed, in which assembly tasks are assigned to specific areas closely related to the place of the performance thereof on the vehicle. The model takes into account additional location restrictions. The proposed approach fits in with current trends aiming at minimizing the space occupied by assembly lines along with reducing the number of stations. A hybrid method is proposed combining modified known heuristics of task assignments to workstations with bounding and backtracking techniques in order to determine the assignment of workers to tasks.
Robotic assembly lines are widely used in manufacturing industries. The robotic assembly line balancing (RALB) problem aims to balance the workloads among different workstations and optimize the assembly line efficiency. This paper addresses a particular type of RALB problem, which minimizes the assembly line cycle time by determining the task and robot assignment in each workstation under precedence constraints. To solve the problem, we present an effective hybrid algorithm fusing the estimation of distribution algorithm and branch-and-bound (B&B) based knowledge. A problem-specific probability model is designed to describe the probabilities of each task being assigned to different workstations. Based on the probability model, an incremental learning method is developed and a sampling mechanism with B&B based knowledge is proposed to generate new feasible solutions. The fuse of B&B based knowledge is able to reduce the search space of EDA while focusing the search on the promising area. To enhance the exploitation ability, a problem-specific local search is developed based on the critical workstation to further improve the quality of elite solutions. The computational complexity of the proposed algorithm is analyzed, and the effectiveness of the B&B based knowledge and the problem-specific local search is demonstrated through numerical experiments. Moreover, the performance of the proposed algorithm is compared with existing algorithms on a set of widely-used benchmark instances. Comparative results demonstrate the effectiveness and efficiency of the proposed algorithm.
In any mechanical manufacturing industry, raw materials’ cost is the major cost component which constitutes up to 40% of the total cost. Other major cost is the processing cost to convert the raw materials in to a finished product that takes place in an assembly line. Optimizing the available resources including the number of work benches and operators is important to meet the timely delivery of finished goods with optimum cost. In an assembly line, the manufacturing processes on any material are sub-divided into many tasks which may include sub-assemblies and machining. These tasks are carried out in several work stations which are responsible for a single or a set of operations carried out on the materials. Assembly lines need to be balanced to have even distribution of work for both men and machines. Type 1 simple assembly line balancing problems (SALBP-1) refer to minimization of number of work stations by keeping the cycle time constant. If the number of work stations is fixed, minimizing the cycle time falls under Type II problems. To solve these types of NP-hard problems, heuristics are used by researchers. The heuristics use different parameters to assign a particular task to a work station. This paper proposes and analyses a few heuristics that can be used to solve SALBP-1. They use slope indices to order the jobs and assign them to different work stations. The slope indices are the ratios of any two parameters taken suitably
Mixed-model assembly lines are large scale production layouts that often operate under uncertainties such as stochastic product sequences. Balancing such lines can be particularly challenging as throughput estimation can be difficult to determine, especially when asynchronous pace and buffers are considered. Recent works have addressed problem variants with a given target throughput, but few authors consider a variant of the throughput maximization of mixed-model assembly line balancing problem. This paper addresses the balancing optimization problem for an assembly line with a given number of workstations and buffers between them. A make-to-order environment is considered, modeled as stochastic sequence of products with known demand rates. A novel specialized cycle time simulator (CTS) is introduced, as well as a simheuristic approach (PSH) that exploits CTS to assess the cycle time of an assembly line and provide good balancing solutions. The proposed simheuristic PSH is applied to a dataset with several buffer layouts, and its solutions are then compared to those of literature benchmarks. Performance comparisons show that PSH’s solutions outperform the benchmarks’ ones, with statistically significant differences. Furthermore, the solution quality difference was greater for instances with more buffers, highlighting PSH capacity to conveniently exploit buffers in assembly lines. Lastly, analyses on the average processing times of stations, obtained for each buffer layout, partially verifies and question established results of the “bowl phenomenon” on unpaced assembly lines.
The literature studies assume that resources used to be perform the tasks are certain and homogenous in any assembly line. However, tasks may need to be processed by general resource requirements in real life. These general resources could be classified by usage of resources such as simple or multiple, alternative and concurrent. The problem which is related to assignment of the task to any workstation and assignment of resources needed by the task simultaneously is defined as resource-constrained assembly line balancing problems (RCALBPs). In this study, a multiobjective model with minimization of cycle time and resource usage for a given number of stations is modeled to solve the RCALBP for the first time. Alternative and general resource types for tasks and using more than two resource type requirements are also considered. A constraint programming model is developed and solved to find the optimal solutions of these problems. The proposed models are tested with sample scenarios to show the effectiveness of the model.
ResearchGate has not been able to resolve any references for this publication.