Manufacturing Complexity: The Effects of Common Attributes of Manufacturing System Design on Performance

ABSTRACTManufacturing operations are complex. In part, this complexity arises from decisions regarding the design of the system, e.g. number of products, breadth of product structure, and number of operations in the routing. Because of the differences in complexity from one operation to another, it is difficult to make comparisons or account for the relative complexity among manufacturers in research. In this article, a large scale simulation of a generic batch-type manufacturing system was conducted to study the impact of eight system design attributes that are common among most manufacturing systems. The results of the study identify the effects of these attributes. Lastly, this article discusses how these effects should be given consideration as managers make decisions about changing a system's design.INTRODUCTIONIf you were to talk to managers of manufacturing plants, you would often hear them something similar to "Our operation is so complex it makes it difficult to manage." Complexity in operations is now being studied more frequently in the literature on management (e.g. Jacobs, 2007; Wu et al., 2002). But what exactly is complexity and what makes systems complex?It is important to study manufacturing complexity. For the practicing manager, understanding the impact of changes in the manufacturing system design can help them improve operation performance. For researchers, being able to account for differences in manufacturing complexity will help to study the impacts of other issues related to business performance.I n this article eight elements of complexity in manufacturing systems attributed to the system's design are studied. In the first part of this article, the literature is explored for a definition for a complex system and we identify several measurable elements of manufacturing complexity resulting from the design of the production system. In the next section, the research design is presented including a description of the large scale simulation of a batch manufacturing system used for this study. Lastly, the results are reported with practical conclusions and suggestions for future research.LITERATURE REVIEWThe Oxford Dictionary defines the term complex as "consisting of many different and connected parts" and "not easy to analyze or understand" ("complex", Oxford Dictionary). In a review of the literature in the areas of physics, general systems theory, philosophy and medicine, there is no single generally accepted definition of complexity (Flood, 1987; Klir, 1985; Lofgren, 1977; Ashby, 1973; Simon, 1962; Stein, 1989). Casti (1979) proposed a good definition stating that a complex system as one that has a counterintuitive, unpredictable or complicated structure and behavior.But, how can the things that make systems counterintuitive and unpredictable, i.e., the complexity, be measured? One approach is to measuring the length of the shortest description of a system (Klir, 1985; Lofgren, 1977; Ashby, 1973; Simon, 1962). The longer the description needed to portray a system, the more complex the system is. However, determining what is the shortest complete description of a system is extremely subjective, thus impracticable.The second approach is to consider the number of elements in the system and the number and type of relationships between these elements (Flood, 1987; Klir, 1985; Lofgren, 1977). This notion can be linked to Simon (1962) who says that a complex system has a large number of parts, whose relationships are not "simple". By considering this approach, the "things" that make a system's description longer are taken into account, thus fulfilling the intention found in the first approach.We recognize that manufacturing systems are complex, because they are unpredictable and often have complicated structures. They are composed of many subsystems and elements, e.g. work centers, machines, components and products. The elements of a manufacturing system have relationships. …