Universität zu Köln

Reiners, Robin ORCID: 0009-0008-8873-9403 (2025). Data Analytics for Inventory Management Demand and Lead Time Estimation for Spare Parts Planning. PhD thesis, Universität zu Köln.

PDF Dissertation_Robin_Reiners.pdf - Published Version Download (1MB)

Abstract

This dissertation investigates data-driven methods for improving spare-parts inventory management under intermittent demand and uncertain lead times. Using ERP master data and purchase orders from a global equipment manufacturer, it first develops machine-learning models to predict procurement lead times more accurately than static ERP settings, reducing safety stocks and inventory costs. Second, it proposes a non-parametric conditional density framework to estimate full predictive distributions of lead-time demand, enabling better service-level control. Third, a simulation study analyzes cross-learning across SKUs, showing when pooling information improves forecasting and when item-level models remain preferable for robust inventory decisions, offering practical guidance for practitioners.

Item Type:	Thesis (PhD thesis)
Creators:	Creators Email ORCID ORCID Put Code Reiners, Robin reiners.robin@icloud.com https://orcid.org/0009-0008-8873-9403 UNSPECIFIED
URN:	urn:nbn:de:hbz:38-793177
Date:	2025
Language:	English
Faculty:	Faculty of Management, Economy and Social Sciences
Divisions:	Faculty of Management, Economics and Social Sciences > Business Administration > Supply Chain Management > Professorship for Business Administration, Supply Chain Management and Management Science
Subjects:	Data processing Computer science Management and auxiliary services
Uncontrolled Keywords:	Keywords Language lead time English machine learning English inventory management English intermittend demand English forecasting English data analytics English
Date of oral exam:	21 November 2025
Referee:	Name Academic Title Thonemann, Ulrich Professor
Refereed:	Yes
URI:	http://kups.ub.uni-koeln.de/id/eprint/79317