Simultaneous minimization of cycle time and workstations with stochastic task times in a continuous assembly line production system

• Main Author: Chan, Juanito S.
• Format: text
• Language: English
• Published: Animo Repository 1997
• Subjects: Stochastic systems; Assembly-line methods; Industrial management; Mathematical models; Line of balance (Management); Scheduling (Management); Industrial Engineering; Industrial Technology
• Online Access: https://animorepository.dlsu.edu.ph/etd_masteral/1824
• Institution: De La Salle University

Based on the literature and articles reviewed, there was not a single study that deals with the simultaneous determination of the cycle time and number of workstations under stochastic times. Most of the considered deterministic task times with only one objective which is minimizing the idle time in stations. While others are concerned with the simultaneous determination of cycle time and number of workstations, the models formulated dealt with deterministic task times as one of their basic assumptions. As a result, this study deals with the simultaneous balancing of cycle time and the number of workstations under the assumption of stochastic task times for the work elements in each station which the previous researches have failed to consider.In this study, the stochasticity of the task times is considered because the time requirement for each task varies with different workers and with repeated performance by the same worker. This is due to the fact that we have also to consider the other various bahavioral aspects of line balancing such as absenteeism, transfer rates and employee turnovers.The greatest impact of considering stochastic task times is to improve the existing line balancing techniques by achieving a better, if not perfect, line balances in a continuous assembly line production system. Furthermore, the optimum solution obtained approximates that of a real world situation. Two Heuristic Procedures were used to solve the stochastic model after solving the deterministic model with the use of the software called LINDO (Linear Interactive Discrete Optimizer). However, a computer program could be developed for the two heuristics given in the study.Based on the results of the study, it concluded that the solutions obtained using both heuristics are the same regardless of the probability used in the model. In addition, the optimum cycle time decreases with an increase in the optimum number of workstations if the set-up cost of each station decreases with constant cycle time cost. However, the line efficiency still remains the same. Furthermore, with very small probability, say as small as 0.0001, the optimum number of workstations increases by 1 and the objective function value also increases.