The implementation of efficient strategic decisions such as process design and capacity investment under uncertainty, during the product development process, is critical for the pharmaceutical industry. However, to tackle these problems the widely used multi-stage/scenario-based optimization formulations are still ineffective, especially for the first-stage (here-and-now) solutions where uncertainty has not yet been revealed.

This study extends the authors’ previous work addressing the stochastic product-launch planning problem, by developing a new Multi-Objective Integer Programming model, embedded in a unified decision-making framework, to obtain the final design strategy that “maximizes” productivity while considering the decision-maker preferences.

An approximation of the efficient Pareto-front is determined, and a subsequent Pareto solutions analysis is made to guide the decision process. The developed approach clearly identifies the process designs and production capacities that “maximize” productivity as well as the most promising solutions region for investment. Moreover, a good balance between investment and capacity allocation was achieved.

在产品开发过程中，执行有效的战略决策（例如过程设计和不确定性下的产能投资）对于制药行业至关重要。但是，为解决这些问题，广泛使用的基于多阶段/基于场景的优化公式仍然无效，尤其是对于尚未揭示不确定性的第一阶段（此处和现在）解决方案。

这项研究通过开发嵌入在统一决策框架中的新的多目标整数编程模型，扩展了作者先前针对随机产品发布计划问题的工作，从而获得了最终设计策略，该策略可以“最大程度地”提高生产力，同时考虑决策者的偏好。

确定有效Pareto前沿的近似值，然后进行后续的Pareto解决方案分析以指导决策过程。已开发的方法清楚地确定了“最大化”生产率的过程设计和生产能力，以及最有希望的投资解决方案领域。此外，在投资和能力分配之间实现了良好的平衡。