Approaches for handling wicked manufacturing system design problems

Abstract Due to complex interdependencies, incomplete information and uncertainty in the manufacturing requirements, manufacturing system design can be described as a wicked problem. A wicked problem is an ill-defined problem, which makes it difficult to determine if a solution has been found and cannot be solved with traditional optimization approaches. This research will present arguments to justify why such manufacturing system design problems can be considered to fall under the definition of wicked problems especially when customer requirements, hence products, evolve over time. Paradigms such as changeable manufacturing, cyber physical and evolvable production systems have emerged to deal with these issues. A critical analysis of these approaches is then presented, which exposes a gap which exists in the design stages of manufacturing systems to adequately support manufacturing system designers in handling the wicked aspect of manufacturing system design problems. This research therefore argues for the importance of developing integrated approaches which consider both business and product evolution aspects when designing manufacturing systems. Finally an integrated approach is contributed which leverages artificial intelligence to support manufacturing system design decision making whilst considering business and product evolutionary aspects.