Header minimieren

Prof. Dr. Christian Mandl

Beschaffung, Logistik, Supply Chain Management
Advanced Analytics, Operations Research, Data Science

Fakultät Angewandte Wirtschaftswissenschaften (School of Management)

Professor:innen

Professor

Leiter des Schwerpunkts “Digitales Beschaffungsmanagement und Supply Chain Management”

A 006.1

0991/3615-5531

Sprechzeiten

Montag, 13:00 - 14:00 (Nach vorheriger Terminvereinbarung via Email)
In Präsenz (Büro A.006.1) oder via Zoom

Publikationen

Sortierung:

Vortrag

Christian Mandl

Von Predictive zu Prescriptive Analytics – Big Data in der Rohstoffbeschaffung

In: DigiCamp der TH Deggendorf

Technische Hochschule Deggendorf Deggendorf

12.04.2022 (2022)

Angewandte Wirtschaftswissenschaften
DIGITAL

Vortrag

Christian Mandl

Prescriptive Analytics for Commodity Storage Applications. Invited Talk.

In: Brown Bag Seminar der Wirtschaftswissenschaftlichen Fakultät

Katholische Universität Eichstätt-Ingolstadt Eichstätt

18.05.2022 (2022)

Angewandte Wirtschaftswissenschaften
DIGITAL

Zeitschriftenartikel

Christian Mandl
S. Nadarajah
S. Minner
S. Gavirneni

Data‐driven storage operations: Cross‐commodity backtest and structured policies

In: Production and Operations Management pg. 1-19.

(2022)

DOI: 10.1111/poms.13683

Abstract anzeigen

Storage assets are critical for physical trading of commodities under volatile prices. State-of-the-art methods for managing storage facilities such as the reoptimization heuristic (RH), which are part of commercial software, approximate a Markov Decision Process (MDP) assuming full information regarding the state and the stochastic commodity price process and hence suffer from informational inconsistencies with observed price data and structural inconsistencies with the true optimal policy, which are both components of generalization error. Focusing on spot trades, we find via an extensive backtest that this error can lead to significantly suboptimal RH policies. We develop a forward-looking data-driven approach (DDA) to learn policies and reduce generalization error. This approach extends standard (backward-looking) DDA in two ways: (i) It represents historical and estimated future profits as functions of features in the training objective, which typically includes only past profits; and (ii) it enforces structural properties of the optimal policy. To elaborate, DDA trains parameters of bang-bang and base-stock policies, respectively, using linear- and mixed-integer programs, thereby extending known DDAs that parameterize decisions as functions of features without policy structure. We backtest the performance of RH and DDA on six major commodities, employing feature selection across data from Reuters, Bloomberg, and other public data sets. DDA can improve RH on real data, with policy structure needed to realize this improvement. Our research advances the state-of-the-art for storage operations and can be extended beyond spot trading to handle generalization error when also including forward trades.

Angewandte Wirtschaftswissenschaften
DIGITAL

Vortrag

Christian Mandl

Optimal Procurement and Inventory Control in Volatile Commodity Markets

In: Annual Conference of the German Society of Operations Research

Bern, Schweiz

31.08.2021 (2021)

Angewandte Wirtschaftswissenschaften

Vortrag

Christian Mandl

Datengetriebene Ansätze zur Optimierung der Rohstoffbeschaffung unter Preisrisiken

Technologie Campus Grafenau Grafenau

08.06.2021 (2021)

Angewandte Wirtschaftswissenschaften

Zeitschriftenartikel

Christian Mandl

Datengetriebene Planungsmethoden zur Steuerung der Rohstoffbeschaffung und des physischen Rohstoffhandels unter Preisrisiko

In: OR News (Das Magazin der Gesellschaft für Operations Research e.V.)

(2020)

Angewandte Wirtschaftswissenschaften
DIGITAL

Beitrag (Sammelband oder Tagungsband)

Christian Mandl

Datengetriebene Ansätze zur Optimierung der Beschaffung und Lagerhaltung von Rohstoffen in volatilen Märkten

In: Supply Management Research. Aktuelle Forschungsergebnisse 2020 (Advanced Studies in Supply Management)

(2020)

DOI: 10.1007/978-3-658-31898-7_3

Abstract anzeigen

Preisschwankungen stellen sowohl für rohstoffverarbeitende als auch für rohstoffhandelnde Unternehmen eine große Herausforderung dar. Die vorliegende Arbeit untersucht die Auswirkungen von Preisunsicherheit auf optimale Beschaffungs- und Lagerhaltungsstrategien. Eine zentrale Erweiterung der existierenden Literatur sind der Fokus auf Preismodellunsicherheit, sprich unvollständige Information über den zugrundeliegenden stochastischen Preisprozess, sowie auf kostenoptimale Beschaffungsentscheidungen statt optimaler Preisprognosen. Die entwickelten stochastischen und datengetriebenen Analytics-Ansätze kombinieren hierbei mathematische Optimierungsverfahren des Operations Research mit Methoden aus dem Bereich des Maschinellen Lernens und liefern als Ergebnis effektive und interpretierbare Entscheidungsregeln. Das primäre Ziel dieser Arbeit ist es, Rohstoffeinkäufern im digitalen Zeitalter Entscheidungsunterstützungstools an die Hand zu geben, mit deren Hilfe Big Data für verbesserte Beschaffungs- und Lagerhaltungsentscheidungen genutzt werden kann. Die entwickelten Algorithmen wurden dabei auf der Basis von Echtdaten für verschiedene Rohstoffklassen (Metalle, Energie, Agrar) validiert.

Angewandte Wirtschaftswissenschaften
DIGITAL

Zeitschriftenartikel

Christian Mandl
S. Minner

Data-Driven Optimization for Commodity Procurement Under Price Uncertainty

In: Manufacturing and Service Operations Management

(2020)

DOI: 10.1287/msom.2020.0890

Abstract anzeigen

Problem definition: We study a practice-motivated multiperiod stochastic commodity procurement problem under price uncertainty with forward and spot purchase options. Existing approaches are based on parametric price models, which inevitably involve price model misspecification and generalization error. Academic/practical relevance: We propose a nonparametric, data-driven approach (DDA) that is consistent with the optimal procurement policy structure but without requiring the a priori specification and estimation of stochastic price processes. In addition to historical prices, DDA is able to leverage real-time feature data, such as economic indicators, in solving the problem. Methodology: This paper provides a framework for prescriptive analytics in dynamic commodity procurement, with optimal purchase policies learned directly from data as functions of features, via mixed integer linear programming (MILP) under cost minimization objectives. Hence, DDA focuses on optimal decisions rather than optimal predictions. Furthermore, we combine optimization with regularization from machine learning (ML) to extract decision-relevant data from noise. Results: Based on numerical experiments and empirical data, we show that there is a significant value of feature data for commodity procurement when procurement policy parameters are learned as functions of features. However, overfitting deteriorates the performance of data-driven solutions, which asks for ML extensions to improve out-of-sample generalization. Compared with an internal best-practice benchmark, DDA generates savings of on average 9.1 million euros per annum (4.33%) for 10 years of backtesting. Managerial implications: A practical benefit of DDA is that it yields simple but optimally structured decision rules that are easy to interpret and easy to operationalize. Furthermore, DDA is generalizable and applicable to many other procurement settings.

Angewandte Wirtschaftswissenschaften
DIGITAL

Hochschulschrift

Christian Mandl

Optimal Procurement and Inventory Control in Volatile Commodity Markets: Advances in Stochastic and Data-Driven Optimization

Technische Universität München München Lehrstuhl für Logistik und Supply Chain Management

15.05.2019 (2019)

Abstract anzeigen

Volatile prices constitute a challenge for both commodity-processing and commodity-trading firms. This thesis investigates the implications of price uncertainty on the optimal operating policies in multi-period procurement and inventory control. A central contribution to the existing literature that addresses the full information problem is the focus on the implications of price model uncertainty, i.e., incomplete information about the underlying price process. Based on advances in stochastic and data-driven optimization, we propose mathematical models for practical decision support and test them on real data. Hence, this thesis gives guidance to managers in the digital age on how to use real-time information and Big Data in combination with methods from statistical learning theory (Bayesian learning, machine learning) in an optimization framework in order to improve commodity procurement and inventory management decisions. The first problem considers operational hedging via inventory control. We show how a Bayesian belief structure can be used to express uncertainty about the price process, which is subject to switches in regimes. We prove the structure of the optimal storage policy and test its cost impact relative to several more practical but suboptimal control policies. We find that Bayesian learning yields significant cost savings. The second problem addresses commodity procurement via forward contracting. We propose a data-driven and machine learning-enabled mixed integer linear programming model that jointly optimizes forecasts and decisions by training optimal purchase signals as functions of features related to the price. Finally, we quantify the performance loss caused by ignoring feature information in procurement. The third problem considers optimal commodity storage from the perspective of a merchant with buying, storing and reselling opportunities. We propose several data-driven models for storage optimization. Based on empirical data of six major exchange-traded commodities, we find that optimally structured data-driven policies can outperform state-of-the-art reoptimization approaches.

Angewandte Wirtschaftswissenschaften
DIGITAL

Zeitschriftenartikel

Christian Mandl
S. Minner

Von Predictive zu Prescriptive Analytics. Big Data in der Rohstoffbeschaffung

In: Beschaffung aktuell

(2017)

Angewandte Wirtschaftswissenschaften
DIGITAL

Kernkompetenzen

Methoden:

Prescriptive Analytics
Datengetriebene Optimierung
Machine Learning

Anwendungen:

Optimierung von Beschaffungsprozessen
Optimierung von Lagerhaltungsprozessen
Optimierung von Logistikprozessen
Optimierung von Lieferketten und -netzwerken

Vita

Unternehmensberater bei McKinsey & Company Inc., München
Gründer von prelytico, München
Forschungaufenthalt an der Cornell University, Ithaca/New York
Promotion am Lehrstuhl Logistik & Supply Chain Management, TU München
M.Sc. Technologie- und Managementorientierte Betriebswirtschaftlehre, TU München
B.Eng. Wirtschaftsingenieurwesen, Hochschule München

Sonstiges

Gutachtertätigkeit

International Journal of Production Economics
OR Spectrum

Preise & Auszeichnungen

Platz 1 bei MSOM Practice-Based Research Competition (zusammen mit Prof. Stefan Minner)
Dissertationspreis der Gesellschaft für Operations Research (GOR)
Wissenschaftspreis des Bundesverbands Materialwirtschaft, Einkauf & Logistik (BME)
Finalist beim Best Student Paper Award der International Society for Inventory Research (ISIR)

Mitgliedschaften

Gesellschaft für Operations Research (GOR)
Bundesverband Materialwirtschaft, Einkauf & Logistik (BME)
TUM Management Alumni e.V.