An integrated approach for the configuration of automated manufacturing systems

The paper proposes a new integrated approach for supporting firms in their decisions of dimensioning automated production systems. The problem is closely related to the performance evaluation of the system since discriminating indicators are necessary to rank different alternatives. Traditionally analytical methods and simulation have been used to evaluate the production system performance, with minor emphasis on the relationships between the tools and their use. Given the complexity of the problem, it is not possible to use only analytical methods that cannot enter deeply in problem details; at the same time the space of potential system configurations is too large to be evaluated by means of detailed tools such as simulation. In the proposed methodology the problem is decomposed hierarchically into different sub-problems; each one has a different level of detail and a specific performance evaluation tool is used. At each level of the analysis each system configuration is compared, by means of statistical tests, with the other alternatives with the purpose of discarding unprofitable solutions.