The Chaos theory and manufacturing at first glance appear to be at best very strange partners. The very word "chaos" infers negative things that no one wants to associate with an efficiently run shop.
The Chaos theory and manufacturing at first glance appear to be at best very strange partners. The very word "chaos" infers negative things that no one wants to associate with an efficiently run shop. For example, the dictionary definition of chaos is great confusion, complete disorder, a confused mass or mixture, the infinite space or formless matter before the universe existed. And because of this negative connotation, there has been a gradual shift to the term Complexity Science to describe Chaos projects. But, ever since the Chaos theory was developed in the 1980s, scientists, researchers, and engineers are reaching a growing awareness that it is offering them new tools for seeing order and pattern where only unpredictable, random, or erratic behavior had been previously observed.
Industry is all too familiar with the complexity of manufacturing systems. In an effort to improve production and quality, meet the market challenges of global competition, build to order, and build it quicker than the competition, manufacturing systems are being pushed to new limits. Older existing sub systems that have been integrated into newer systems have created enormous and very complex software configurations which are often not very robust. This has caused many to take a look at solving the problem of complexity using the Chaos theory.
The Chaos theory refers to processes that can be defined by nonlinear equations. Also, the theory of Chaos indicates that success can never be achieved by a top-down, command and control approach. Such systems become so intertwined that they are nearly impossible to change. Chaos instead advocates simpler rules applied to components of a process, which in turn come together to create a robust, agile system. These components or "autonomous agents" have behaviors that emerge as the system runs rather than being defined in advance and then rapidly executed as in the top-down system.
When the shop floor is viewed as networks of individual agents, each responding autonomously to local conditions, the factory is viewed not as a single machine but as a community of loosely coupled processors. Each of these processors or entities has its own agenda to allow agents to respond locally to changing conditions. Changing to this structure requires management to change from the traditional role of specifying the top level schedule activities to specifying the local decision rules that agents use to schedule their own operation.
Chaos was first applied to manufacturing in 1991 by R. Morley, Inc. (RMI) to seven truck body painting booths at General Motors. Each booth became an autonomous control agent that individually selected its next task by bid from a conveyor line of mixed body styles and color specifications. The results minimized paint changes in several booths saving time and materials. The simple system replaced a simulation that ran more than 500 pages of computer code by reducing it to just a few pages of computer code.
Researchers have used similar concepts of Chaos as tools to optimize simple computer code and engineering designs, and to control complex industrial systems so it is only natural that they should be extended to the shop floor.
There are at least four organizations with programs working to bring about the integration of Chaos and Complexity Science into industry:
Early expectations of Chaos and Complexity Science are high. This is a field of research that merits watching closely in the next few years to see if it can provide the necessary tools needed for shops to meet the manufacturing challenges of the 21st century. In application, chaos becomes much less theory and more reality.