EAGER: A Decision Analytic Framework for Large-Scale Design and Manufacturing

EAGER：大规模设计和制造的决策分析框架

• 批准号: 1565168
• 负责人: Ali Abbas
• 金额: $ 13.5万
• 依托单位: University of Southern California
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1565168&HistoricalAwards=false
• 关键词: EAGER; Decision; Analytic; Framework; Large

The research objective of this EArly Concept Grant for Exploratory Research (EAGER) award is to derive a multidisciplinary theory of normative decision making for design and manufacturing organizations. The theory will provide new axioms for normative decision making for two types of enterprises: profit maximizing firms (such as John Deere) and federal/governmental design enterprises (such as NASA Langley Research Center). The theory will translate the axioms into mathematical formulations for consistent decision making for the ultimate purpose of improving the design and manufacturing decisions within the United States. For a profit maximizing firm, a fundamental axiom will be investigated: that there be no division within an organization that will accept a project if another division within the same organization, and with the same information and resources, would reject. This axiom will translate into the necessity of a corporate utility function: divisions within the same organization will operate with the same risk attitude when making design decisions. The research will illustrate how to assess a corporate utility function and will explore the decision making behavior that results from common incentive structures within design firms. The work will also derive unique incentive structures that align incentivized behavior with expected utility maximizing behavior. Behavioral experiments will be conducted to verify whether the incentive structures are feasible and match the desired expected utility framework. Verification of the results will also be achieved through industry collaborators. A new Bayesian framework for demand estimation will also be explored and will be used in maximizing the expected utility of profit. The research will include a significant educational component with NASA Langley, John Deere, and the engineering design community. If successful, the research will advance the theory and practice of engineering design by introducing several new theoretical foundations for normative decision making within the design process. The theory will (i) align design requirements with the value function for the firm, (ii) align incentive structures with expected utility maximizing behavior, (iii) provide axioms needed for consistent decision making within a design firm. The high risk elements within this award include verifying the new methods with the industry partners and NASA, as well as identifying the transition mechanism to incorporate the decision framework within the current design organizations. Improving the decision making within design enterprises is a cornerstone of the U.S. economy. By appropriately identifying the processes needed for consistent decision making, individuals, design engineers, and various business enterprises will produce better products. Besides the theoretical and experimental work, this award will enable the PI to conduct two workshops in collaboration with NASA Langley, John Deere and other members of the design and manufacturing community: the first will be on "Setting Value-Based Requirements for Engineering Design" and the second will be on 'Aligning Incentive Structures with Expected Utility Maximization'. The research will be disseminated broadly and the award will enable the PI to continue a joint collaboration the Champaign County Juvenile Detention Center (JDC) spreading decision skills to troubled teens.

这一早期的探索性研究概念补助金(EIGER)奖的研究目标是为设计和制造组织推导出一种标准化决策的多学科理论。该理论将为两类企业的规范决策提供新的公理：利润最大化企业(如John Deere)和联邦/政府设计企业(如NASA兰利研究中心)。该理论将把这些公理转化为数学公式，以做出一致的决策，最终目的是改善美国国内的设计和制造决策。对于一家利润最大化的公司，我们将研究一个基本公理：如果同一组织内具有相同信息和资源的另一个部门拒绝接受一个项目，则该组织内部不存在接受该项目的部门。这一公理将转化为公司效用函数的必要性：同一组织内的部门在做出设计决策时将以相同的风险态度运作。这项研究将说明如何评估企业效用函数，并将探索设计公司内部常见激励结构导致的决策行为。这项工作还将得出独特的激励结构，使受激励的行为与预期的效用最大化行为保持一致。将进行行为实验，以验证激励结构是否可行，以及是否符合期望的期望效用框架。对结果的验证也将通过行业合作伙伴实现。还将探索一种新的贝叶斯需求估计框架，并将用于最大限度地实现利润的预期效用。这项研究将包括与NASA兰利、约翰·迪尔和工程设计社区的重要教育组成部分。如果成功，这项研究将为设计过程中的规范决策引入几个新的理论基础，从而推动工程设计的理论和实践。该理论将(I)使设计要求与公司的价值函数相一致，(Ii)使激励结构与预期效用最大化行为相一致，(Iii)为设计公司内部一致的决策提供所需的公理。该奖项中的高风险因素包括与行业合作伙伴和NASA验证新方法，以及确定将决策框架纳入当前设计组织的过渡机制。改善设计企业内部的决策是美国经济的基石。通过适当地确定一致决策所需的流程，个人、设计工程师和各种商业企业将生产出更好的产品。除了理论和实验工作外，该奖项还将使PI能够与NASA兰利、John Deere和其他设计和制造社区成员合作举办两个研讨会：第一个研讨会将是“为工程设计设定基于价值的要求”，第二个研讨会将是“使激励结构与预期效用最大化保持一致”。这项研究将被广泛传播，该奖项将使PI能够继续与香槟郡青少年拘留中心(JDC)合作，向问题青少年传播决策技能。