A heuristic algorithm to approximate dynamic program of a novel new product development process

详细信息    查看全文

• 作者：Hamed Fazlollahtabar^a ; ^{hfazl@iust.ac.ir" class="auth_mail" title="E-mail the corresponding author} ; Iraj Mahdavi^b ; Nezam Mahdavi-Amiri^c
• 关键词：New product development (NPD) ; Dynamic programming ; Heuristic algorithm
• 刊名：Journal of King Saud University - Engineering Sciences
• 出版年：2016
• 出版时间：January 2016
• 年：2016
• 卷：28
• 期：1
• 页码：84-91
• 全文大小：1650 K

文摘

We are concerned with a new product development (NPD) network in digital environment in which the aim is to find integrated attributes for value added purposes. Different views exist for new product development. Here, the effective factors are categorized into customers, competitors and the company’s own past experience. Also, various attributes are considered for the development of a product. Thus, using digital data of attributes, the optimal set of attributes is chosen for user in the new product development. Regarding the multi stage decision making process of the customer, competitor and company’s own past experience, we develop a multi-dimensional dynamic program as a useful tool for multi stage decision making. To counteract the dynamism of the digital data in different time periods, two concepts of state and policy direction are introduced to determine the cost of moving through the stages of the proposed NPD digital network. Since the space requirements and value function computations become impractical for even moderate size, we approximate the optimal value function developing a heuristic algorithm.