In this paper, we propose a method for determining the operating condition of process variables to satisfy given quality requirements in process industries such as chemical, metallurgical, iron industry and so forth. Under the condition that the quality characteristics satisfy the requirements, the method provides a desirable level of an objective function, defined as a linear combination of process variables. The method requires that data of quality characteristics and process variables have been collected. Also, a linear regression model is assumed to be the functional relationship betwen quality characteristics and process variables.First, from the viewpoint of customer satifaction, it is necessary to determine an operating condition assuring that the probability Q(x) for quality characteristics satisfying the requirements is higher than specified level q and then two criteria for desirable operating condition are derived.Second, the region of operating condition is derived, which provides Q(x)≥q with confidence coefficient 100(1-α)% by constructing a β-content tolerance interval. Finally, a heuristic solution for desirable level providing on the objective function is obtained through the joint use of the Kuhn-Tucker Condition Method and the Parametric Method from the field of mathematical programming.

View full abstract

[in Japanese]

View full abstract