Georghios P. Sphicas

Deterministic Equivalents for Stochastic Assembly Line Balancing

Abstract Two methods for considering variations in task completion times in the assembly line balancing problem are (1) assign tasks to stations such that the sum of the completion times is less than a fixed proportion of the cycle time and (2) assign tasks to stations such that the probability for completing all tasks within the cycle time is greater than a fixed value. In this paper, we show that for certain classes of distributions of task completion times, the first method, which is basically a deterministic approach to the problem, is equivalent to the second, which requires a stochastic formulation. The nature of the relationship is also demonstrated.

A systematic approach to determine the optimal maintenance policy for an automated manufacturing system

We propose a systematic approach to determine the optimal maintenance policy for an automated manufacturing system which includes a flexible manufacturing cell (FMC) and several automated machine shops. The systematic approach combines simulation, fractional factorial design, noise or outer array of Taguchi design, regression metamodelling, and classical queueing analysis. A useful expression of the fractional utilization of the manufacturing system is derived and incorporated into formulating and solving the corresponding decision problem. The systematic approach provides an effective implementation procedure to handle practical maintenance problems found in a complex manufacturing environment.

The Optimal Aiming Line

Track and field athletes have traditionally pursued performance optimization in an artistic rather than a scientific manner. But many of the tactical decisions made by athletes in these events, most of which are currently made by intuition, can be analysed quantitatively to provide optimal results for the participants.

An evaluation of alternative task allocations to two servers

The problem considered here is how to allocate n exponential tasks to two available servers. The servers may be placed in a series or parallel configuration, and arrivals are assumed to be Poisson. In general, this gives rise to a network of queues with Erlang service times. Alternative designs are examined using various approximation methods and simulation for several values of n and different utilization rates. If the workers must be placed in tandem, the rule of balancing the work of the two stations is found to be optimal for all cases considered. If it is possible to place the servers in a parallel configuration, each performing all n tasks, the average total waiting time can be substantially reduced.

Design of Inspection Policy in an Assembly Oriented Flexible Manufacturing System

Abstract Since zero-defect technologies would be elusive and costly in any manufacturing system, the in-process quality control is necessary in a flexible manufacturing system (FMS) in practice. Hence, an analytical model is proposed to describe the operational charcteristics of the FMS with quality control. There are three major contributions of this paper. First, an open queueing network with general time distribution is introduced to bridge the gap between the quality control and the FMS. Second, the issues of part feedback, reworking and scrapping are considered simultaneously in an unreliable FMS which has not been studied before. Finally, a mathematical model is presented to find an approximate solution of random inspection rate.