Russell E. King

Manufacturing cell design: an integer programming model employing genetic algorithms

The design of a cellular manufacturing system requires that a part population, at least minimally described by its use of process technology (part/machine incidence matrix), be partitioned into part families and that the associated plant equipment be partitioned into machine cells. At the highest level, the objective is to form a set of completely autonomous units such that inter-cell movement of parts is minimized. We present an integer program that is solved using a genetic algorithm (GA) to assist in the design of cellular manufacturing systems. The formulation uses a unique representation scheme for individuals (part/machine partitions) that reduces the size of the cell formation problem and increases the scale of problems that can be solved. This approach offers improved design flexibility by allowing a variety of evaluation functions to be employed and by incorporating design constraints during cell formation. The effectiveness of the GA approach is demonstrated on several problems from the literature.

INTEGRATING KANBAN TYPE PULL SYSTEMS AND MRP TYPE PUSH SYSTEMS: INSIGHTS FROM A MARKOVIAN MODEL

Approaches to multistage production scheduling can be conveniently classified into push type (i.e., Materials Requirements Planning (MRP) systems) or pull type (i.e., kanban systems). Each is generally thought to have both advantages and disadvantages. In this paper, a hybrid push/pull strategy is developed with the aim of gaining the advantages of both approaches. Material flow between work centers is regulated using the standard single card kanban/pull arrangement. Superimposed on this is the MRP-type information flow which feeds forward demand information directly to one or more (but not necessarily all) work centers. A general N-stage hybrid push/pull model is developed. The use of the approach is illustrated using 3-stage and 4-stage serial flowlines. The results indicate that the push/pull approach has lower inventory levels and a better response to demand changes than the pure pull system. The hybrid approach seems to combine many of die advantages of MRP approaches while retaining much of the simp...

Retail performance measures for seasonal fashion

Retail success can be defined as achieving high gross margins and customer service levels (i.e. being in‐stock) with as little inventory as possible. Forecast accuracy, process lead‐time, offshore/local sourcing mix and up‐front/replenishment buying mix can have a significant impact on success in connection with sourcing seasonal products with a fashion content. Forecast accuracy depends on the characteristics of the product and supply lead‐time. Lead‐times are traditionally long and buying decisions are often made seven to eight months prior to the start of the selling season. Forecast errors lead to some of the items being liquidated at clearance prices while others stockout and lead to lost sales. As a result retailers often resort to higher mark‐up prices with fashion products. However, typical retail performance measures such as service level, lost sales, product substitute percentage, gross margin, gross margin return on inventory, sell‐through percentage and mark‐down rate mask the source of the problems. In this paper, we discuss these performance measures and propose a new one. Additionally, case study analysis of a group of Finnish department stores is presented.

Scheduling with alternatives: a link between process planning and scheduling

The objective of this research is to develop and evaluate effective, computationally efficient procedures for scheduling jobs in a large-scale manufacturing system involving, for example, over 1000 jobs and over 100 machines. The main performance measure is maximum lateness; and a useful lower bound on maximum lateness is derived from a relaxed scheduling problem in which preemption of jobs is based on the latest finish time of each job at each machine. To construct a production schedule that minimizes maximum lateness, an iterative simulation-based scheduling algorithm operates as follows: (a) job queuing times observed at each machine in the previous simulation iteration are used to compute a refined estimate of the effective due date (slack) for each job at each machine; and (b) in the current simulation iteration, jobs are dispatched at each machine in order of increasing slack. Iterations of the scheduling algorithm terminate when the lower bound on maximum lateness is achieved or the iteration limit is reached. This scheduling algorithm is implemented in Virtual Factory, a Windows-based software package. The performance of Virtual Factory is demonstrated in a suite of randomly generated test problems as well as in a large furniture manufacturing facility. To further reduce maximum lateness, a second scheduling algorithm also incorporates a tabu search procedure that identifies process plans with alternative operations and routings for jobs. This enhancement yields improved schedules that minimize manufacturing costs while satisfying job due dates. An extensive experimental performance evaluation indicates that in a broad range of industrial settings, the second scheduling algorithm can rapidly identify optimal or nearly optimal schedules.

A Stochastic Model of the Apparel-retailing Process for Seasonal Apparel

A computer model is described that simulates the seasonal apparel-retailing process. The model is stochastic in nature and is designed to allow the investigation of the effects of improved retailing procedures on financial and other performance measures. Its principal value lies in the evaluation or Quick Response (QR) supply methodologies that allow frequent re-estimations of consumer demand and reorders of merchandise based on in-season point-of-sale (POS) data at the stock-keeping-unit (SKU) level.

A review of automated guided-vehicle systems design and scheduling

Abstract Automated guided-vehicle (AGV) systems are gaining increasing acceptance in modern manufacturing facilities primarily because of the versatility they offer. As the flexibility and complexity of these systems increase, the requirements of the design effort and the routeing and scheduling system grow. In this paper we provide a review of the literature relevant to the system design, routeing and scheduling, and justification and implementation of AGV systems.

Satisfying due-dates in large multi-factory supply chains

A procedure is developed for the simultaneous scheduling of multi-factory supply chains, including inter-factory transportation. A job-shop scheduling procedure, known to provide near-optimal solutions to industrial-sized problems, is enhanced to include transportation elements in the fundamental model. In order to demonstrate the quality of the solutions, a lower bound calculation is compared to the procedures solutions on a number of large-scale test problems. The lower bound is an enhancement of the classic lower bound calculation for the N-job, M-machine job shop. The computational effort in scheduling is linear in the size of the problem, and high quality solutions to large-scale problems can be obtained in seconds.

An Apparel-supply System for QR Retailing

A novel apparel-supply system is described that is compatible with Quick Response retailing of apparel with a finite shelf life. The system is driven by a retail point-of-sale procedure, which regularly re-estimates customer demand and generates frequent reorders on the manufacturer and fabric supplier. The system is shown to come close to perfect supply over a range of operating conditions and thus allow greatly improved retail performance when compared with traditional retailing procedures.

A hybrid genetic algorithm for manufacturing cell design

Abstract Global competition is demanding innovative ways of achieving manufacturing flexibility and reduced costs. One approach is through cellular manufacturing, an implementation of the concepts of group technology. The design of a cellular manufacturing system requires that a part population be at least minimally described by its use of process technology (padmachine incidence matrix) and partitioned into part families and that the associated plant equipment be partitioned into machine cells. At the highest level, the objective is to form a set of completely autonomous units such that inter-cell movement of parts is minimized. This paper presents a stochastic global optimization technique utilizing genetic algorithms (GAS) and local improvement procedures (LIPs) to solve the cell design problem. The combination of LIPs with GAS is shown to improve the performance of the GA in terms of solution quality and computational efficiency. Several different methods of incorporating these procedures into the GA are investigated. The concepts used in these hybrid techniques can easily be extended to other variations of the cell design problem as well as to other LIPs.

ROBOT TASK SCHEDULING IN A FLEXIBLE MANUFACTURING CELL

Effective sequencing and scheduling of the material handling system can have a major impact on the productivity of the manufacturing system. This is especially true in the case where material handling times are on par with machine processing times. In a dynamic, real-time environment, the optimal solution of this scheduling problem may be computationally infeasible. In this paper, we develop a branch and bound approach which is coupled with quick, effective bounds to optimize the movement of a robot which serves the material handling requirements within a manufacturing cell. Computational results are given which explore the tradeoff between computation time and deviation from optimal for different scenarios.