The operation and evaluation of flexible manufacturing systems
Abstract
Flexible Manufacturing Systems (FMSs) combine machining centers and automatic work stations connected by automatic material handling devices, with all operation under computer control. Their objective is to increase the efficiency and productivity of batch production by providing the advantages of automation to the production of a large variety of different part-types whose limited, individual quantities could never justify automatic mass manufacturing. For the efficient on-line control of material flow in an FMS this research proposes a look-ahead procedure of product routing utilizing the on-line information provided by the computerized system. After a product completes an operation, the procedure determines the next machine the part should join, by considering the next few parts which will be ready for the next operation. In this way, product flow time and product processing time are reduced. While the material flow in an FMS is highly synchronized, most existing formulae for estimating product flow congestion use only the means and variances of underlying distributions of the queueing system, and are not specific to the effect of scheduling. This research develops new formulae which possibly reflects some of these effects. The formulae also use the data which are easier to get in a manufacturing system and quantify some modeling assumption errors. This research also addresses a vital issue of FMS operation: the estimation of labor cost reduction from a new technology. In the past, reduced labor costs were usually estimated in a deterministic way, assuming that estimation of demand rates was accurate and the reduction in labor costs was proportional to the reduction in labor hours. However, when there is demand uncertainty this often introduces bias in the estimation. This research develops a split and merge procedure for estimating the distribution of reduction in labor hours obtained from a new investment under uncertain demand.
Degree
Ph.D.
Advisors
Liu, Purdue University.
Subject Area
Industrial engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.