Sensing the Future: Designing Sensor-based Predictive Information Systems for Forecasting Spare Part Demand for Diesel Engines
Year: 2018
Type: dissertation
Abstract: As digital technologies become prevalent and embedded in the\nenvironment, "smart" everyday objects like smart phones and smart homes\nhave become part and parcel of the human enterprise. The ubiquity of\nsmart objects, that produce ever-growing streams of data, presents both\nchallenges and opportunities. In this dissertation, I argue that information\nsystems extending these data streams, referred to as "predictive\ninformation systems with sensors", can generate added value and will be\ngaining momentum in academia and in the industry. Subsequently, seeing\napparent complexity in designing IS artifacts with such functionality, I\nintroduce a framework for Designing Information Systems with Predictive\nAnalytics (DISPA), extending Design Science Research specifically towards\nrigorous design of predictive analytics.\nThe framework is evaluated based on a case study of MAN Diesel and\nTurbo, a lead designer of marine diesel engines generating multiple\napplicable artifacts in the process. Additionally, the framework\nexemplification in the case context led to supplementing the framework\nwith a set of Design Principles for Designing Predictive Information\nSystems as well as a matrix for pre-assessing financial feasibility of\npredictive information systems with sensor technologies.\nThis work provides a contribution to information systems research, and in\nparticular to design science research, by introducing a model for Designing\nInformation Systems with Predictive Analytics (DISPA) that can serve as a\nmethod for developing IS artifacts. The framework constitutes an\nInformation System Design Theory consistent with the established\ndefinitions from the literature (Gregor & Jones, 2007; Kuechler & Vaishnavi,\n2012; Walls, Widmeyer, & El Sawy, 1992). In addition, the paper introduces\nand systematically evaluates a number of spare-part forecasting methods,\nwhich can be considered a contribution to operations research literature.
Source: Econstor (Econstor)
Author: Simon Furtak
Institution: Copenhagen Business School
Language: English
Subfield: Management Information Systems
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W3092810907