Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/5427
Title: ANALYSIS AND OPTIMISATION OF FLOW LINE PRODUCTION SYSTEMS
Authors: Hira, Dilbagh Singh
Keywords: MECHANICAL & INDUSTRIAL ENGINEERING
RANDOM VARIATION
FLOW CHARACTERSTIC
FLOW LINE PRODUCTION SYSTEMS
Issue Date: 1983
Abstract: A flow line production system is characterised by a number of stages arranged in series, where the work material passeg through all the stages in a fixed sequence. The varia-bility in the production rates of the stages cause t; delays and hence loss in production. A number of factors, external and internal to the system, may be responsible for the variability in production rates of the stages. Two factors; which have generally been considered in the literature are : (i) Random variations in operation times of the stages. (ii) Breakdowns of the stages. It has been found that in most of the existing litera-ture, only one of these two causes of production line ineffi-ciency has been considered. In each case the exact analytical models of line efficiency are available for two or three stage lines only, The application of numerical methods is also limit-ed to small problems only. The empirical models, developed from the simulation data, have been found to be specific in nature and limited in application. A little work has been done to analyse the lines having variable operation times, as well as subject to breakdowns. Also, not much work has been done to study the performance of unbalanced lines. In this thesis, simulation, the most versatile tool for analysing the large complex systems, has been employed. Flow line systems having ; (v) • The applicability of the empirical models has been demonstrated with the help of illustrative examples and a case study. In each case the sensitivity analysis has been perform-ed to examine the effect of system parameters on the optimum size of the inprocess buffers. The sensitivity of the net gain from the system, to deviations from the optimum values of in-process buffer size and the number of repairmen has also been examined. The subject matter of the thesis has been presented under the following heads : 1, Introduction 2. Review of past work 3. System description and design of simulation experiments 4, Balanced flow lines with random operation times 5. Unbalanced flow lines with random operation times 6. Balanced flow lines with unreliable stages 7. Unbalanced flow lines with unreliable stages S. Flow lines with random operation times and unreliable stages 9. Effect of repair policy and crew size on line performance 10. Conclusions and scope for future work, and Bibliography (i) Random operation times (manual flow lines), (ii) Unreliable stages (automatic transfer lines), and (iii) Random operation times as well as unreliable stages, have been simulated in FORTRAN IV to study the influence of system parameters on the line efficiency and the work -in - process inventory. Validity of the simulation models has been tested by comparing the simulation results with the available data reported by others.. Simulation experiments have been designed to ensure a reasonable degree of confidence in the results. The data thus obtained has been analysed to develop simple and reliable empirical models for the efficiency of balanced manual flow lines, as well as, for the balanced automatic transfer lines. While the empirical models in case of manual flow lines are valid for normal and Erlang distributions of operation times of the stages, the empirical model for the efficiency of automatic transfer lines is applicable when the failure and repair times are exponential. In each case the cost analysis of the system has been conducted and expressions for the optimum inprocess buffer size derived.
URI: http://hdl.handle.net/123456789/5427
Other Identifiers: Ph.D
Appears in Collections:DOCTORAL THESES (MIED)

Files in This Item:
File Description SizeFormat 
TH MIED 177740.pdf10.89 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.