This study presents an algorithm for deriving the long-term polices of quality level, price and advertisement for a product. The diVusion models and cost functions are combined to formulate pro® t functions capable of determining future pro® t trends. The algorithm ® rst implements the optimal control theory to derive the optimal conditions of the pro® t function. Then the genetic algorithm is employed to search for the approximate solutions of quality level, price and advertising expenditure at each period on the planning horizon (life cycle). Examples of diVerent scenarios of the model parameters are presented to describe the results obtained herein. Sensitivity analysis for the major parameters is performed to specify their eVects on pro® ts. Results in this study allow us, ® rstly, to obtain explicit solutions simultaneously with respect to quality, price and advertising policies, secondly, to propose an appropriate algorithm for solving the diVerent scenarios of the dynamic pro® t function, which consists of the diVusion function and cost function, and thirdly, to enhance the long-term pro® t performance via the polices proposed herein, that is the approximation of the best solution.
NomenclatureA…t † rate of advertising expenditure at time t C…x…t †; q…t † † total cost per unit at time t for a cumulative sales volume x…t † and quality level q…t † d price parameter h quality parameter M total number of potential purchasers over the life cycle of the product P…t † unit price at time t q…t † quality level at time t, 0 < q…t † 4 1 r discount rate T period of the life cycle x…t † cumulative sales volume by time t x 0 …t †ˆg…x…t †; p…t †; q…t †; A…t † †, sales rate at time t for a cumulative sales volume x…t †, price p…t †, quality level q…t † and advertising expenditure A…t † ¬ innovation coe cient reaction coe cient for advertising ® imitation coe cient
Maintenance float problems in a flexible manufacturing system tend to analyti cally intractable while the assumptions of their models become more complex and prac tical; consequently, they are too complex to solve by traditional optimization theories; and therefore, a heuristic algorithm or an enumerative search method is always used to solve them. To tackle the pr'Jblems, in this study, we solve them by a genetic algorithm which is often implemented to search the global optimization in a complex . s.ea.rch space. The numerical example shows that our approach can find the optimal solution of these problems with less effort.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.