Layout design in fractal organizations

Today's complex, unpredictable and unstable marketplace requires flexible manufacturing systems capable of cost-effective high variety–low volume production in frequently changing product demand and mix. In fractal organizations, system flexibility and responsiveness are achieved by allocating all manufacturing resources into multifunctional cells that are capable of processing a wide variety of products. In this paper, various fractal cell configuration methods for different system design objectives and constraints are proposed. These parameters determine the level of interaction between the cells, the distribution of different product types among the cells and the similarity of cell capabilities. A tabu-search-based method is proposed to optimize the product distribution to the cells and the arrangement of machines and cells on the shop floor. This optimization is performed for different fractal cell configuration methods and cell quantities. The quality of the resulting shop floor layouts is measured in terms of resource requirements and material movements. The results indicate that in fractal layouts, a trade-off is required between machine quantities and material travelling distance. It was generally possible to reduce travelling distances by increasing the degree of optimization on machine layout and product distribution for a specific product demand and mix.     

A design methodology for fractal layout organization

Venkatadri el a.(HE Transactions, 29, 911-924, 1997) have proposed a new methodology for shop floor layout that involves the use of fractal cells and have compared the performance of their new layout with those obtained using the function, group and holographic layouts. A few inconsistencies are present in their results, expressed as flow scores. This note points out these inconsistencies through the use of appropriate examples.     

Capability-based distributed layout approach for virtual manufacturing cells

It is shown in the literature that in highly volatile manufacturing environments functional job shops and classical cellular manufacturing systems do not perform well. Classical cellular manufacturing systems are very sensitive to changing production requirements due to their limited flexibility. In order to adapt cellular manufacturing systems to volatile manufacturing environments, the virtual cellular manufacturing concept was proposed in the 1980s by the National Bureau of Standards in USA. This concept is similar to group technology where job families are processed in manufacturing cells. The main difference between a virtual cell and the classic cell is in the dynamic nature of the virtual manufacturing cell; whereas the physical location and identity of classic cell is fixed, the virtual cell is not fixed and will vary with changing production requirements. The virtual manufacturing cell concept allows the flexible reconfiguration of shop floors in response to changing requirements. In the literature, the formation and scheduling process of virtual cells are clearly explained and researched in detail. However, the layout issue is not addressed entirely. Virtual cells are generally formed over functionally divided job shops. Forming virtual cells over a functional layout may adversely affect the performance of a virtual cellular manufacturing system. There is a need to search for different layout strategies in order to enhance the performance. The distributed layout approach may be a better alternative for virtual cellular manufacturing applications. In this research paper, a novel capability-based approach is proposed for the design of distributed layouts. A simulated annealing based heuristic algorithm is developed from the distributed layout. The proposed approach is tested with a problem with real data. An example is also shown in order to give an idea about the superiority of a capability-based distributed layout over the functional layouts in forming virtual manufacturing cells.     

Design of flexible plant layouts

In this paper, we present an approach for the design of plant layouts in stochastic environments. We consider systems where the, product mix and product demand are subject to variability and where duplicates of the same department type may exist in the same facility. In contrast to a job shop layout, we allow these duplicates to be placed in non-adjacent locations on the plant floor and for flow allocation between pairs of individual departments to be made as a function of the layout and the product demand realization. We present a scenario-based procedure that iteratively solves for layout and flow allocation. We show that having duplicates of the same departments, which can be strategically located in different areas of the plant floor, can significantly reduce material handling cost while effectively hedging against fluctuations in flow patterns and volumes. We show that the effect of duplication is of the diminishing kind, with most of the cost reduction occurring with relatively few duplicates. We also show that the quality of the obtained layouts can be quite insensitive to inaccuracies in estimating demand scenario probabilities.     

Simulation study of coordinating layout change and quality improvement for adapting job shop manufacturing to CONWIP control

Constant work-in-process (CONWIP) has been highlighted for its superiority over kanban for application in job shop environments. Smooth operation of CONWIP yet depends on some critical practices, including: layout change (from functional to cellular), quality improvement (QI) and set-up time reduction induced by layout change (STR). Simulation modelling was employed in this study to investigate the coordination of these conditions. QI was experimented with the mean magnitude and probability of occurrence for step shifts (MMSS and POSS, respectively); it has major effects on flow time variability only. If STR is below 60%, a functional layout with push control was found favourable while QI is not effective. STR needs to reach 60% to justify layout change; POSS and MMSS reductions are effective for cellular layouts with CONWIP and push control, respectively. At this stage, promotion of improvement activities should be the major concern for replacing push control with CONWIP. The performance of CONWIP is superior if a 70% or larger STR is achieved. In this case, it enables greater delivery performance upgrade via strengthening the effectiveness of QI in reducing flow time variability. Essential guidelines were ultimately derived for reforming traditional job shop practices, progressing through layout change, towards the realisation of CONWIP.     

Learning and labour assignment in a dual resource constrained cellular shop

Recent studies have shown that in a cellular shop that benefits from learning due to repetitive processing, limitations attributable to routing flexibility can be more than offset. Moreover, the shop can respond more quickly to changes in demand than a job shop. This study examines the impact of labour assignments in a dual constrained cellular shop in which processing times decrease with operator task repetition. Results indicate that in the presence of operator learning, shop performance is significantly affected by the flexibility permitted in labour assignments. Moreover, the sensitivity of performance to labour assignments is significantly impacted by staffing levels and the magnitude of learning effects.     

No-wait two-machine permutation flow shop scheduling problem with learning effect,common due date and controllable job processing times

We consider a two-machine no-wait permutation flow shop common due date assignment scheduling problem where the processing time of a job is given as a function of its position in the sequence and its amount of resource allocated to this job. The common due date (CON) assignment method means that all the jobs are given a common due date. We need to make a decision on the common due date, resource allocation and the sequence of jobs to minimise total earliness, tardiness, common due date cost and total resource cost. We show that the problem remains polynomially solvable under the proposed model.     

The integrated machine allocation and layout problem

Cellular manufacturing has received a considerable amount of attention in the research literature as an approach for improving the performance of manufacturing facilities. However, recent studies have shown that cellular layouts are not always superior to the traditional functional machine layout. We propose a model that does not require the machines to be placed in a functional layout or in a cellular arrangement, but allows the material flow requirements to dictate the machine placement. The model is formulated as an aggregation of the quadratic assignment problem and several network flow problems coupled with linear side constraints. A mixed integer program is presented to find the optimal solution for small problems, and heuristics are developed to solve larger problems. Computational results evaluating the quality of the solution methodologies are also presented.     

Minimizing work-in-process and material handling in the facilities layout problem

We consider the plant layout problem for a job shop environment. This problem is generally treated as the quadratic assignment problem with the objective of minimizing material handling costs. Here we investigate the relationship between material handling costs and average work-in-process. Under restrictive assumptions, an open queueing network model can be used to show that the problem of minimizing work-in-process reduces to the quadratic assignment problem. In this paper, we generalize these results through a simulation model, and develop a simple secondary measure which allows us to select the layout that minimizes average work-in-process levels from among solutions that are similar with respect to the objective function for the quadratic assignment problem.     

Performance comparison of virtual cellular manufacturing with functional and cellular layouts in DRC settings

This study investigates the performance of virtual cellular manufacturing (VCM) systems, comparing them with functional layouts (FL) and traditional, physical cellular layout (CL), in a dual-resource-constrained (DRC) system context. VCM systems employ logical cells, retaining the process layouts of job shops. Part family-based scheduling rules are applied to exploit the benefits of group technology while retaining the flexibility and functional synergies of the job shop. Past studies of VCM have been based entirely on single-resource-constrained (SRC) systems, i.e. as purely machine-limited systems, assuming that resources such as labour and tooling do not restrict the output. However, given the fact that labour forms a second major constraining resource, and many of the advantages associated with cellular manufacturing are derived from labour flexibility, it becomes necessary to extend the research to DRC systems. In this study, we assume several levels of labour flexibility in all three systems, in addition to other relevant factors such as lot size, set-up reduction, and labour assignment rules. It is shown that VCM can outperform efficiently operated FL and CL in certain parameter ranges, as preliminary research has shown so far. However, it is shown that CL tends to outperform both VCM and FL in the parameter ranges customarily advocated for CL, namely, low lot sizes, adequate levels of set-up reduction, cross training of workers, and worker mobility within cells.     

Tooling constraints and shop floor scheduling: evaluating the impact of sequence dependency

In a previous study, we began the process of examining how finite tooling resources affect the operation of a simulated job shop. The paper extends this line of research by focusing on the problem of how to schedule a shop operating with both finite machine and tooling resources, and faced by varying levels of tooling-related sequence dependency in the setups. Four tool assignment rules and two dispatching procedures are evaluated. Findings show the importance of using tool assignment rules which consider not only tool-related information such as tool availability but also the priority of the jobs waiting to be processed. Specifically, as the level of sequence dependence increases and the availability of tooling decreases, the performance of the shop is greatly influenced by the tool assignment rule used to manage the flow of tooling to and from work centres     

Optimal due-date assignment in a job shop†

This paper presents an analytical model for determining the optimal processing-time and number of operations multiples for the TWK and TWK + NOP due-date assignment methods in a dynamic job shop subject to restrictive assumptions on queue discipline and processing time distribution. The analytical results are compared with the experimental results obtained from simulation of a hypothetical job shop under various shop conditions. The close agreement of the results reveals the validity of the analytical model. In addition, the results show that the TWK + NOP method is more effective in minimizing the missed due-date cost in a job shop.     

A combined dispatching criteria approach to scheduling dual flow-shops

This study investigates a dual flow-shops scheduling problem. In the scheduling context, there are two flow shops and each shop involves three processing stages. The two shops are functionally identical but their stage processing times for a job are different. While sequentially going through the three processing stages, each job is allowed to travel between the two shops. That is, for a job, each of its three stages could be processed in any of the two shops. Such a context is called dual flow-shops in the sense that the two flow shops’ capacities are completely shared. The scheduling problem involves two decisions: (1) route assignment (i.e. assigning the processing stages of a job to a shop), and (2) job sequencing (i.e. sequencing the jobs waiting before each stage). The scheduling objective is to minimise the coefficient of variation of slack time (CV_S ), in which the slack time (also called lateness) denotes the difference between the due date and total completion time of a job. We propose five genetic-algorithm-based (GA-based) solution methods to solve the scheduling problem, which are called GA-EDD, GA-FIFO, GA-SPT, GA-LFO, and GA-COMBO respectively. Numerical experiments indicate that GA-COMBO outperforms the other four methods.     

A hybrid approach for concurrent layout design of cells and their flow paths in a tree configuration

The layout design of multiple-cell automated manufacturing systems includes cell layout design and flow path layout design. Traditional layout methods often treat these two as separate problems and the sequence for solving them is usually cell layout first and flow path layout later. However, approaches of these kinds have one major drawback, that is, they may produce cell layouts that are awkward or difficult for designers to conduct flow path layouts, or cell layouts that do not turn out to be as good as expected after flow path layouts have been performed. Other drawbacks of traditional layout methods include irregular shapes of cells, inaccurate calculations of flow distances, etc. This paper addresses the layout problem of cells and their connecting flow paths in a tree configuration. The proposed layout procedure is designed to avoid the aforementioned drawbacks of traditional layout methods by emphasizing concurrent layout design of cells and flow paths. It combines a search algorithm and mathematical programming models. The search algorithm has a backtracking procedure that allows one to explore alternative layouts, while the mathematical programming models help one obtain accurate layouts of cells and flow paths. The proposed layout procedure also interacts with designers and allows designers to include their qualitative consideration into the layout design. As a result, one can obtain more accurate and good-quality layouts with the proposed layout procedure.     

A simulation comparison of group technology with traditional job shop manufacturing

A simulation model of an actual job shop was used to compare group technology with traditional job shop manufacturing. The experiment compared shops which had four different layouts, designed to emphasize different features of traditional job shops and group technology shops, and four distributions of demand for end items. The group technology shops exhibited superior performance in terms of average move time and average set-up time. The traditional job shops had superior performance in queue related variables (average queue length, average waiting time, work-in-process inventory, etc.). This was caused by group technology's dedication of machines. The effects of the queue related variables outweighed the effects of average move time and average set-up time: the average flow time was shorter in the traditional job shop than in the group technology shops.     

Information and scheduling in a dual resource constrained job shop

In today's job shop, computers make information readily available, but there is limited research about which information to use and how to use it controlling a dual resource constrained (DRC) job shop. We develop and test new order review/ release (ORR) job dispatching and labour assignment rules that use different types of information. We compare the performance of a naive ORR rule (i.e., releasing all jobs immediately upon receipt) and other rules in the literature which use little information to a new ORR rule, modified load conversion (MLC), which uses all the available information. Our tests demonstrate that MLC performs better than the other ORR rules we tested. But the new job dispatching and labour assignment rules developed to use different types of information do not significantly improve performance. A sensitivity analysis demonstrates that although the shop performance is sensitive to the cost structure for every policy tested, the new MLC ORR rule performs well over a wide range of cost structures.     

Family-based scheduling of shops with functional layouts

Simulation studies of job shop scheduling have typically assumed that either setup times are zero (subsumed within the processing time), or that every part has such a unique setup that no setup advantages can be gained by better scheduling policies. These studies also assume that the shop has exactly one copy of every machine. Some researchers have proposed heuristics that explicitly consider setup times and parallel machines in the context of a one stage shop with static arrivals. In contrast, family-based scheduling centred around setup time reduction has been credited with achieving economic savings in batch production industries where GT is employed. We motivate this study by the case of an existing realworld semi-conductor testing facility that has family setups, parallel machines and dynamic job arrival. Using this setting, we investigate whether benefits can still be obtained by using a family-based scheduling philosophy in those environments which do not permit the physical creation of cellular layouts due to the presence of process related or other constraints. We propose and evaluate two new dispatching procedures in a functional job shop that is modelled after the semiconductor testing facility. Results show that a family-based scheduling philosophy centred around coordinating machine setups is advantageous at relatively high setup to processing time ratios, while classic job shop rules suffice otherwise. Based on these results, we present recommendations for managing such environments. We also suggest future research directions in this area.     

Optimal allocation of resources in a job shop environment

In this paper we study the allocation of production services (e.g., maintenance) to machining centers in a job shop when there is a limited amount of resources for such services. Different classes of jobs go through the shop. A job class is characterized by its route, its processing requirements, and its priority. The problem we address is how to optimally allocate production services (the resources) to machining centers so as to minimize the total Work-In-Process in the entire system. In order to analyze this problem we model the job shop as an open queueing network. Assuming certain relationships between performances of machining centers (i.e., speeds) and the amounts of resources allocated, we present methods for allocating the resources to the individual machining centers optimally.     

Investigating the factors influencing the shop performance in a job shop environment with kanban-based production control

In this paper, we intended to explore three research questions related to applying just-in-time (JIT) manufacturing techniques in a job shop environment with the kanban-based production control. Simulation experiments were performed and the results were analysed using a statistical method, planned comparison. We found that the adoption of cellular layout in a job shop with the kanban system should only be considered if the amount of setup time reduction achievable through cellular manufacturing is medium to large; otherwise, functional layout should be adopted. In addition, moving parts in batches within cells was found to be more advantageous than moving parts one piece at a time. Only minor effects of material handling speed on the shop performance were identified.     

Efficient heuristic approaches to transform job shops into flow shops

In this paper, we address the problem of transforming a job shop layout into a flow shop with the objective of minimizing the length of the resulting flow line. Since this problem is NP-hard, we focus our attention on developing high quality approximate solutions. We start by reviewing existing heuristics for the problem as well as some heuristics developed for the Shortest Common Supersequence problem, a well-known stringology problem similar to the one under consideration. We then present a new decomposition approach for the problem that allows the application of local search techniques. We have embedded this approach into a tabu search procedure that is shown to be effective in subsequent computational experiments. Finally, we provide best-so-far solutions for classical job shop problem instances, so they can be used as benchmark instances for further research.