Exploratory Research On Engineering The Service Sector: Towards A Decision Informatics Paradigm

这项研究将寻求机会，将以工程和科学为导向的工具和方法推广到新兴类型的服务企业。本着这一目的，该研究在考虑了美国服务业的范围、规模和独特特征后，从工程角度确定了新兴服务业的特征，这反过来预示着将建立一个决策信息学范式，以加强服务的生产和提供。在这一新的批判性范式框架内，研究重点是多个非同质数据源的实时融合和分析，从两个批判性角度强调了该研究的智力价值。首先，它是问题驱动的，而不是解决方案驱动的。因此，新兴服务的工程相关特征清楚地确定了对决策信息学的需求，这代表着向实时、信息驱动的决策的范式转变。其次，它将大大增加关于多个数据源的实时分析和融合的知识。虽然有相当多的关于数据挖掘和模式分析的文献，但缺乏进行实时数据分析的工具。例如，从实时决策的角度来看，关键是要知道最近接收到的数据点是数据点历史分布的一部分，还是实际上是新现象的开始的离群值--因此，自适应的实时决策不仅受数据输入的影响，还受输入的顺序的影响。此外，虽然有相当多的文献关于融合来自多个同质--主要是定量--来源的数据，但缺乏对来自多个非同质(即定量和定性)来源的数据进行实时融合的工具。此外，多个非同质数据源的同时实时融合和分析是一个没有得到调查的领域；然而，新兴服务--特别是电子(电子)服务--需要这种工具来加强其生产和交付。还应指出的是，该研究力求产生最广泛的影响，特别是在研究与教育的结合方面。首先，它构成了博士论文工作的基础。其次，开发的实时数据融合和分析工具将应用于现有的服务问题，使用来自多个非同质来源的真实数据。第三，成果将在多个场馆展示和发布。第四，研究结果将被适当地组织起来，并整合到由首席调查员教授的两门现有课程中；一门是名为“决策评估方法”的研究生课程，另一门是名为“以决策为中心的系统工程”的本科课程。第五，根据这项研究的探索性性质，将确定更多的研究需要，以进一步巩固决策信息学方法，类似于生物信息学的持续发展。

This study will pursue opportunities for extending engineering- and science-oriented tools and approaches to newly emerging types of service enterprises." Consistent with this purpose, the study, after considering the scope, size and unique features of the U.S. service sector, identifies characteristics of emerging services from an engineering perspective, which, in turn, augur for a decision informatics paradigm for enhancing the production and delivery of services. Within the framework of this new and critical paradigm, the study is focused on the real-time fusion and analysis of multiple non-homogeneous data sources.The intellectual merit of the study is underscored from two critical perspectives. First, it is problem-driven, not solution-driven. Thus, it is the engineering-related characteristics of emerging services that have clearly identified a need for decision informatics, which represents a paradigm shift towards real-time, information-driven decision making. Second, it will add significantly to the body of knowledge on the real-time analysis and fusion of multiple data sources. While there is considerable literature on data mining and pattern analysis, tools for undertaking real-time data analysis are lacking. For example, from a real-time decision perspective, it is critical to know whether the most recently received data point is part of the historical distribution of data points or an outlier that is, in effect, the beginning of a new phenomenon -- thus, adaptive, real-time decision making is not only affected by data input but also by the sequence of that input. Additionally, while there is considerable literature on the fusion of data from multiple, homogenous -- mostly quantitative -- sources, tools for undertaking real-time fusion of data from multiple, non-homogeneous (i.e., quantitative and qualitative) sources are lacking. Moreover, the coincident real-time fusion and analysis of multiple, non-homogeneous data sources represents an area that has eluded investigation; yet, emerging services -- especially e(lectronic)-services -- require such tools to enhance their production and delivery. It should also be noted that the study seeks to have the broadest impact, especially in regard to the integration of research and education. First, it constitutes the basis for a doctoral dissertation effort. Second, the developed, real-time data fusion and analysis tools will be applied to an existing service problem, employing real data from multiple, non-homogeneous sources. Third, the results will be presented and published in a number of venues. Fourth, the results will be appropriately structured and integrated into two existing courses being taught by the Principal Investigator; a graduate course, entitled "Evaluation Methods for Decision Making," and an undergraduate course entitled "Decision-Focused Systems Engineering." Fifth, consistent with the exploratory nature of the study, additional research needs will be identified to further underpin the decision informatics approach, similar to the continuing development of bioinformatics.