|
Supplier selection with AI-based TCO models: Cost prediction case study in an automotive OEM
Spreitzenbarth, Jan
|
Dokumenttyp:
|
Konferenzveröffentlichung
|
|
Erscheinungsjahr:
|
2021
|
|
Buchtitel:
|
65th AACE international Conference & Expo 2021
|
|
Seitenbereich:
|
1362-1384
|
|
Veranstaltungstitel:
|
65th AACE International Conference & Expo 2021
|
|
Veranstaltungsort:
|
Online
|
|
Veranstaltungsdatum:
|
14.-18.06.2021
|
|
Ort der Veröffentlichung:
|
Red Hook, NY
|
|
Verlag:
|
Curran Associates
|
|
ISBN:
|
978-1-7138-3365-9
|
|
Verwandte URLs:
|
|
|
Sprache der Veröffentlichung:
|
Englisch
|
|
Einrichtung:
|
Fakultät für Betriebswirtschaftslehre > Stiftungslehrstuhl für Procurement (Bode 2014-)
|
|
Fachgebiet:
|
004 Informatik
330 Wirtschaft
|
|
Abstract:
|
The goal of this research is to understand more clearly the lifecycle costs of supplier selection using methods of artificial intelligence (AI) with a total cost of ownership (TCO) model to reduce uncertainty and make better decisions. AI is a key technology for operations management and its usage is still in its infancy. Few have successfully integrated AI methods into their operations and across their supply chains but are recently starting to emerge. The research is driven by the question of how to reduce uncertainty to provide better information for selecting the right supplier. A case study is conducted at a German automotive manufacturer based on three interlinked data sets. These include: 1. Naïve algorithm models are evaluated as baselines for quality of cost prediction based on supplier selection nomination. 2. Engineering and production changes are analyzed since they often lead to price increases. 3. Cost breakdowns are considered, as they are applicable during several lifecycle phases. For the last 50 years, AACE International and the project management community have made significant contributions to increase the maturity in the practice of project management and control. This continuous commitment applies to remain resilient in the era of data science. This study suggests practical ways to break down uncertainty into a measurable quantity. References are drawn from the Total Cost Management Framework and the applicability is discussed to other settings such as construction, aerospace, defense, and public procurement where considerable related research is conducted. The work confirms previous research that in particular regression trees and Bayesian optimization can reduce the uncertainty inherent in supplier selection more than previously utilized methods.
|
 | Dieser Eintrag ist Teil der Universitätsbibliographie. |
 Suche Autoren in
Sie haben einen Fehler gefunden? Teilen Sie uns Ihren Korrekturwunsch bitte hier mit: E-Mail
Actions (login required)
 |
Eintrag anzeigen |
|
|
