Fuzzy multistate system analysis and design using imprecise customer information

In order to become successful in an intense competitive world, companies have to incorporate system level customer requirements in the design stage in a manner that help design the components. System level requirements include the expected state of the system with a degree (membership function) that the system belongs to this state. These system level requirements are allocated to the component level by minimizing the cost when the system and its components are composed of multiple states and the states are fuzzy (states are not discrete).; The mathematical framework based upon fuzzy concepts and the steps involved in designing system with fuzzy states are presented. The states of the system are defined by the combinations of the fuzzy states of the components called component state vectors. The membership function of the system state (degree that the system belongs to a certain state) is derived based upon the membership functions of the states of the components in the component state vector. The concept of functional characteristic is used to determine the membership functions of the states of the components.; Using ∝-cut approach the fuzzy state of the system is derived from the states of the components in the component state vector. Thus, the mathematical framework based upon fuzzy concepts developed in this research characterizes the multistate system.; The allocation process of the system state requirements is described by searching for combination(s) of states of the components and their membership functions at minimum cost points that minimize total cost. A computer program in visual basic is developed to ease the process of allocation.; At the end, an example using gasoline engine cooling system is presented in order to demonstrate the allocation process.