Fractional cell formation in group technology using modified ART1 neural networks

The primary objective of group technology (GT) is to enhance the productivity in the batch manufacturing environment. The GT cell formation problem is solved using modified binary adaptive resonance theory networks known as ART1. The input to the modified ART1 is a machine-part incidence matrix comprised of the binary digits “0” and “1”. And the outputs are the list of part families and the corresponding part list, machine cells and their corresponding list of machines, and the number of exceptional elements. This method is applied to the known benchmarked problems found in the literature and it is found to outperform other algorithms in terms of minimizing the number of the exceptional elements. The relative merits of using this method with respect to other known algorithms/heuristics in terms of computational speed and consistency are presented .

This paper considers the machine-part clustering under the assumption of alternative process plans for each part. Kusiak's p-median model of part family formation dealing with the problem has critical disadvantages in that the model requires too many binary variables and constraints and the number of part families must be known in advance. Furthermore, the solution quality of the model in terms of the number of exceptional elements is poor for ill-structured problems. Motivated by Viswanathan's work, this paper proposes two new p- median models using new measures of similarity between machine pairs: one with the prespecified number of cells and the other without the prespecified number of cells. Computational experience shows the applicability of new p-median models of machine cell formation over Kusiak's model.

The aim of this chapter is to propose a new heuristic for Machine Part Cell Formation problem. The Machine Part Cell Formation problem is the important step in the design of a Cellular Manufacturing system. The objective is to identify part families and machine groups and consequently to form manufacturing cells with respect to minimizing the number of exceptional elements and maximizing the grouping efficacy. The proposed algorithm is based on a hybrid algorithm that combines a Variable Neighborhood Search heuristic with the Estimation of Distribution Algorithm. Computational results are presented and show that this approach is competitive and even outperforms existing solution procedures proposed in the literature.

This paper addresses the cell formation problem in group technology (GT). We develop two heuristic methods for generating solutions to the problem. These methods are based on two powerful combinatorial search methods simulated annealing and tabu search. The performance of the heuristics is examined using randomly generated, published and industry data. The results indicate that the simulated annealing based heuristic is the preferred technique in the context of the problem addressed in this paper. Further, we also demonstrate that the simulated annealing based heuristic generates near-optimal solutions to the cell formation model formulated in this paper.

An efficient nonhierarchical clustering algorithm, based on initial seeds obtained from the assignment method, for finding part-families and machine cells for group technology (GT) is presented. By a process of alternate clustering and generating seeds from rows and columns, the zero-one machine-component incidence matrix was block-diagonalized with the aim of minimizing exceptional elements (intercell movements) and blanks (machine idling). The algorithm is compared with the existing nonhierarchical clustering method and is found to yield favourable results.

A capacity constrained mathematical programming model for cellular manufacturing with exceptional elements

The Group technology (GT) constitutes a manufacturing philosophy that exploits similarities in product design and product processes. The information attained from the process sheets are organized in an incidence matrix with machines and parts. The goal is to assign parts to families and machine to cells, which are designed to produce a given part family such that the number of voids and exceptional elements in cells are minimized. This article considers the problem of the manufacturing cell formation, of combinatorial nature and proposes a hybrid genetic algorithm (GA) with a greedy cosntrutive method for its solution, aiming the minimizing of the inter-cell movement and maximizing the use of the machines inside a cell. Basically, the GA generates sets of machine cells, and the constructor method is applied to those cells to better assign part families to them. The k-means algorithm is also applied to refine these formations. The performance of the proposed framework, considering the efficacy of grouping and a set of GT problems available in the literature, is presented and discussed.

Cell formation in group technology: Review, evaluation and directions for future research

Cell Formation CF is regarded as critical aspect of a Cellular Manufacturing System CMS. It focuses on the creation of part families and machine cells. There are numerous methods, algorithms and models suggested to deal with solving the CF problem. However, there are several drawbacks to some existing CF methods such as, their inability deal with large size datasets, ill structured datasets, the simultaneous creation of clusters of parts and machines, lack of ability to find optimum solutions, and increased material handling costs. Thus, the suggested polar coordinates effectively address the CF problem by clustering parts and machines to reduce the material handling costs. It offers efficient initial clustering, and adapts to different production environments. While not globally optimal, it provides good starting solutions for further optimization, excelling in cost reduction and efficiency. A polar coordinates-based approach combines the repetition of parts and machines with angles and then the resulting pattern is used for checking similarity criteria. The purpose is to improve the Grouping Efficacy GC, decrease the travel time of parts inside and outside cells and enhance Machine Utilization MU. GC is applied as a performance measure to identify the quality of the suggested CF approach. After the polar-based approach is applied, it enhanced the proportion of exceptional elements and machine utilization. On the other hand, it showed better results for GC when compared with Rank Order Clustering (ROC). For some datasets, it recorded (60, 90) compared with (53.33, 71.93) for the traditional ROC

Cellular Manufacturing (CM) is an important application of Group Technology (GT) in which families of parts are produced in manufacturing cells or a group of various machines, which are physically close together and can entirely process a part family. The manufacturing system established based on such an idea is called Cellular Manufacturing System (CMS). A major problem associated with many CMSs is the existence of Exceptional Elements (EEs), i.e. bottleneck machines and exceptional parts. These are machines/parts that cannot be exclusively assigned to a machine cell/part family. In this paper a new model is presented for dealing with the EEs in the form of a Multi-objective Optimization Problem (MOP). This model aims to minimize: (1) intercellular parts movements, (2) total cost needed for machine duplication and part subcontracting, (3) the system's under-utilization, and (4) deviations among the cells' utilization. Attaining an ideal solution, which is optimal to all of the objectives is prohibited, as they conflict with each other. Hence, a Multi-Objective Genetic Algorithm (MOGA) is developed to provide the decision-maker with a set of non-dominated or Pareto-optimal solutions. Comparisons between the developed MOGA and three other MOGAs show its viability in three performance aspects, namely: quality, diversity and CPU time.

Modeling of a manufacturing cell design problem with fuzzy multi-objective parametric programming

Group technology (GT) is a manufacturing philosophy which identifies and exploits the similarity of parts and processes in design and manufacturing. A specific application of GT is cellular manufacturing (CM). The first step in the preliminary stage of cellular manufacturing system (CMS) design is cell formation, generally known as a machine-part cell formation (MPCF) or a machinecomponent grouping (MCG) problem. Simulated annealing (SA) is not only a highly effective and general random search method to obtain near-global optimal solutions for optimization problems, but also quite appropriate for the MPCF problem which is an NP complete, complex problem. In this study, we introduce modified SA with the merits of a genetic algorithm (GA), call parallel SA (PSA), and propose a PSA-based procedure to solve the MPCF problem. More specifically, this study aims to minimize (1) total cost which includes intercell and intracell part transportation cost and machine investment cost, (2) intracell machine loading unbalance and (3) intercell machine loading unbalance under many realistic considerations. The illustrative example, comparisons and analysis demonstrate the effectiveness of this procedure. The proposed procedure is extremely adaptive, flexible, efficient and can be used to solve real MPCF problems in factories by providing a robust manufacturing cell formation in a short execution time.

The cellular manufacturing system is a well-known strategy which improves the production efficiency and cutting down the manufacturing cost. The main problem in designing cellular manufacturing system is cell formation, which is concerned with grouping parts with similar processing required into part families and associated machines into machines and minimises the exceptional elements. This paper presents the artificial immune system method to solve the manufacturing cell formation problems with the objective to minimise the exceptional elements. This method is applied to the existing datasets found in the literature and it is found to out performs other methods.

Purpose: Use mathematical models of Mixed Integer Linear Programming oriented to cellular distribution and aggregate production planning in order to obtain the appropriate product family for each manufacturing cell and from this, minimize production and material handling costs through the allocation of production resources.Design/methodology/approach: This article develops two mathematical models in LINGO 18.0 software, performing the computational calculation to obtain the best efficiency in cell formation at minimum production cost.Findings: The mathematical model oriented to the formation of manufacturing cells allows a grouping of products and machines with 82.5% group efficiency. By reallocating machines to each cell and redistributing facilities, the cost of material handling is reduced by 35.1%, and the distance traveled in product manufacturing is reduced by 26.6%. The mathematical model of aggregated planning provides information on production resource requirements such as personnel, machinery, distances traveled, as well as the cost generated by the need to outsource part of the production, inventory maintenance and overtime work.Research limitations/implications: It is necessary to clearly define the capacity variables. The model does not take into account the cost of mobilizing machines and readjusting facilities.Practical implications: The case study company can adequately plan production and efficiently manage its resources.Social implications: The study can be applied to other textile SMEs.Originality/value: The aggregate production planning model requires the assignment of the mathematical model of manufacturing cell formation in order to calculate the resource requirements needed to meet a demand.

This paper presents an electromagnetism-like mechanism (EM) algorithm in order to solve cell formation problems. Based on the attraction-repulsion principle in electromagnetic theory, each combination of the part and machine groups was regarded as a charged particle. The total force calculation and the move along the total force calculation were then determined out to either maximize the cell group effectiveness or minimize the exceptional elements (EE) and the number of vacancies (NV). A comparison with other algorithms in the related literature found that the final result of the EM algorithm met the expected quality in terms of the exceptional elements and number of vacancies, regardless of whether or not new cells were added. Key words: Cellular manufacturing system, group technology, electromagnetic-like mechanism.