Developing an optimization model to determine AGV fleet size given the capacity of machines and vehicles in the production industry

Sammanfattning: Background: In the present competitive environment manufacturing firms have shifted their production from mass production to mass customization. In this line, a flexible manufacturing system (or FMS) has evolved to the changing requirements. The estimation of fleet size of AGVs has shown to be a critical decision to increase the efficiency of material handling systems by increasing throughput and reduce delay in manufacturing.[4] The initial investment cost and the total life cycle cost of AGV systems rely on the number of AGVs required.   Objectives: The objective of this study is to analyze the optimal fleet sizing of AGVs in terms of the capacity needs of machines/stations and the available capacity of the AGVs working in the production site. The aim of this study is to give potential users of AGVs an optimization model to consider when planning the optimal number of AGVs needed given their capacity needs.   Methods: The methodology presents our contribution to a general model for deciding fleet size with multiple AGV’s defined as a “big system”. In comparison to previous models, we consider how the capacity of AGV’s can change depending on several factors leading to congestion and delay. The study defines these factors, finds the cause behind them and categorizes them to be used as input variables for the model. The methodology explains the important factors needed for implementing the general model for specific cases when conducting the investigation. The final part of the methodology will discuss the reliability and validity of our approach.  Results: the result presents how the general model was applied at the manufacturing company of Emballator Växjöplast AB. The data gathered and information from the capacity analysis were used to measure the demand for transportation, the AGV capacity, and the factors affecting delay and congestion. The values were then used in the model to determine the optimal fleet size for the case. The result gave a high indication for accuracy and that the general model can be used in specific cases. Finally, we conducted an analysis of how delays and congestion affected the supply of transportation as fleet size increases. The result indicated that the optimal fleet size of AGVs that satisfy 35 machines/stations was 1,32 while the theoretical fleet size of AGVs was 1.3. This yielded a result of 98.84% accuracy in estimating the optimal fleet sizing of AGVs in terms of the capacity needs of machines/stations and the available capacity of the AGVs working in the production site. Conclusions: The result shows that the approach of analyzing the capacity needs of the production site and the capacity available to the AGV can accurately be used to estimate the optimal fleet sizing of AGVs. The implication of this study and the optimization model that considers capacity needs and capacity available rather than specific layout characteristics will allow users to cope with the changing requirements of mass customization. The users will consider their demand forecast and use the optimization model to help them plan the optimal fleet sizing of AGVs.