Optimal sequential and parallel prototyping policies

Summary form only given. When designing a new product or service, its developers must balance the expected profit of improving the design against the costs and time requirements of doing so. The authors propose to quantify this trade-off by modeling a latter stage of development, prototyping, as a search process. The decision makers choose how many prototypes to build, then observe the projected profit of each. Multiple parallel prototypes are modeled as n random draws from a continuous probability distribution. The model balances the cost of building and testing the parallel prototypes against improvements in expected profits. The authors also study the option to abandon the project when even the best prototype generates losses. Further, they investigate prototyping decisions for products consisting of independent components that must be compatible in order for the total system to be profitable. Process design also considers alternatives for improving particular process steps, but is complicated by the need for compatibility between these improved process steps.