Optimization of Supply Chain Management by means of Artificial Intelligence and Machine Learning

Potentials and Advantages of Machine Learning

Because of progressing digitization of supply chains, more and more data, which can be evaluated beneficially, is being generated and can lead to valuable findings. In this context the term artificial intelligence (AI) is also often used. As a part of AI, machine learning (ML) used within supply chain management offers high potentials and many advantages.

What are the Fields of Application within Supply Chain Management?

The possible uses of ML-procedures within supply chain management are multifaced. In almost every field of activity in SCM you can find cases of application: Starting with sales forecast and consulting area information ML is an important part of future work. Further fields of application extend from procurement, e.g. over an ML-based stock- and service level management with dynamic stock disposition, over production with subjects such as e.g. Smart Maintenance to distribution of goods and products, e.g. with consideration in route planning.

Fields of application, in which machine learning is an important tool
© Fraunhofer IML

ML in Sales Planning and Demand Forecasting

The quality of a demand forecast has a significant influence on the quality of following planning processes. As internal data as well as those factors which are possibly influencing the demands can be identified and adequately considered while using external data, there exist an enormous potential in approving supply chains compared to the status quo with help of ML-based methods.

How to Implement ML-based Demand Forecasting

For the most accurate forecast possible of future demand figures, you need different data in different forms and structures as well as actual forecasting methods. Those form the centrepiece of the ML-application and – like the data - need to be determined according to each application case. However, it is possible that in certain use-cases conventional, statistic and therefore non-learning algorithms are more suitable than ML-procedures. The individual requirements for each use-case should therefore always be checked – there is none general-purpose solution!

The various effects of using more or less data as well as using ML-procedures or a statistic procedure, are shown by our ML-demonstrator. We help you to increase your forecasting quality with state-of-the-art-technologies!

To Help You Get Started with Machine Learning, We Offer the Following Services:

machine learning in seven business units
© Fraunhofer IML

Creating an understanding for ML-Usage in your company

  • Definition of individual goals and requirements

Data inspection

  • Inspection of available data
  • Determination of possible problems with data quality
  • Consulting regarding further relevant data sources

Data preparation

  • Data curation, how to handle missing or flawed data
  • Data reduction and transformation
  • Construction of final records for modeling


  • Analysis of the effects when using a specific process
  • Selection of the best ML-/statistic-process
  • Determination of several models


  • Selection of the best model for the use case


  • Preparation and presentation of the results
  • Advice for further action


  • Training of employees in collection and preparation of data as well as utilization of ML-based toolsets for demand planning

Prototypical Realisation of our Works on the Basis of our Machine Learning Demonstrator

The ML Demontrator

On the one hand the prototypical realisation of our machine learning demonstrator shows the feasibility, on the other hand it demonstrates potentials and necessities of data preparation as well as an analysis of demand influencing factors.


Our Publications on Machine Learning

Things to know about Machine Learning in 10 points

Bauen Sie sich Ihre ML-Experten auf
© Fraunhofer IML
Wie bei konventionellen statistischen Verfahren der Bedarfsplanung erfordert die Kalibrierung der ML-Verfahren methodisches Knowhow und ist ein kontinuierlicher Optimierungsprozess.

1. Build your team of ML experts!

Like in conventional, statistical methods of demand planning, the calibration of ML-methods demands methodical knowhow and is a continuous optimization process.

2. Examine and optimize your data situation!

To gain valid findings out of data with an algorithm, a critical amount of data in a proper quality has to be available. We advise a potential evaluation to clarify your data situation.

Prüfen und optimieren Sie Ihre Datenlage!
© Fraunhofer IML
Damit ein Algorithmus aus Daten valide Erkenntnisse ziehen kann, muss eine kritische Menge an Daten in der richtigen Qualität verfügbar sein. Lassen Sie eine Potentialbewertung durchführen, um zu klären, wie gut Sie diesbezüglich aufgestellt sind.
Entwickeln Sie ein umfassendes Datenverständnis!
© Fraunhofer IML
Nur wer den Inhalt seiner Daten kennt und versteht, kann dem ML-Algorithmus sinnvolle und vollständige Datensätze bereitstellen - eine zentrale Voraussetzung für den Erfolg dieser Verfahren.

3. Develop a comprehensive understanding of data!

Only those who understand and know the content of their data can provide meaningful and complete data sets for the ML algorithm – a central requirement for the success of this methods.

4. Consider the individual requirements!

There is none general-purpose-method. In fact, an ML-algorithm has to be selected reasonably. Considerable general conditions are for example the demand pattern, the necessary forecast horizon, the available data base and the accepted calculation effort.

Berücksichtigen Sie die individuellen Anforderungen
© Fraunhofer IML
Ein ML-Algorithmus muss anhand des Anwendungsfalls sinnvoll ausgewählt werden. Zu berücksichtigende Rahmenbedingungen sind hier z.B. das Bedarfsmuster, der erforderliche Prognosehorizont, die verfügbare Datenbasis und der akzeptierte Berechnungsaufwand.
Berücksichtigen Sie den Aufwand der Datenaufbereitung!
© Fraunhofer IML

5. Consider the effort of data preparation!

For being able to train ML-approaches you need to prepare data with regard to the application case with high time expenditure. Experience shows, that more than half of the total effort goes into data collection, data adjustment and -organising, formation of training data sets etc.

6. Bring in domain knowledge!

In order to implement an ML-based algorithm successfully, a close exchange between specialists and methodologists is needed. Build an ML-Team to unite the knowledge of sales- and ML-experts.

ML-Team für gebündeltes Fachwissen
© Fraunhofer IML
Um eine ML-basierte Prognose erfolgreich in die Anwendung zu bringen, müssen Fach- und Methodenspezialisten in engem Austausch stehen. Bilden Sie ein ML-Team, um das Wissen Ihrer Vertriebs- und ML-Experten zusammenzuführen.
Halten Sie Ihr ML-Team auf dem Laufenden!
© Fraunhofer IML
Um die Leistung der ML-Algorithmen zu verbessern, ist ein intensives Feature Engineering erforderlich. Hierbei werden die Daten so aufbereitet, dass der Algorithmus mögliche Korrelationen zwischen einer Einflussgröße (bspw. Marketingaktion) und der Ergebnisgröße (bspw. Absatzmenge) überhaupt erst erkennen kann. Sorgen Sie daher dafür, dass neue Entwicklungen aus dem Tagesgeschäft in Ihrem ML-Team ankommen!

7. Keep your ML-Team up to date!

An intense feature engineering is necessary in order to increase the efficiency of ML-algorithms. This provides a certain preparation of data, so that the algorithm is able to recognise possible correlations between influencing factors (e.g. marketing efforts) and the result quantity (e.g. sales amount). Therefore, make sure that new developments from daily businesses get the ML-teams attention!

8. Train!

The longer a model is trained, the more exact it is able to include characteristics of data sets.

Modelle müssen mit Datensätzen trainiert werden.
© Fraunhofer IML
Je länger ein Modell trainiert wird, desto genauer kann es die Charakteristika des Datensatzes berücksichtigen.
Nutzen Sie die neuen Möglichkeiten!
© Fraunhofer IML
Insbesondere für die Bedarfsprognose existieren inzwischen vielfältige ML-basierte Verfahren und Erfahrungswerte. Tauschen Sie sich mit anderen Unternehmen über Ihre Erfahrungen in diesem Bereich aus und binden Sie externe Experten mit ein.

9. Use new possibilities!

Meanwhile there exist diverse ML-based methods and experience values especially for demand forecast. Share your experiences within this sector with other companies and include external experts.

10. ML is no self-purpose: Use the best method!

Companies, customers and products are different from each other. Therefore, you need to analyse which method leads to the best forecast qualities. Sometimes conventional, statistic methods show better results than the latest AI-algorithms.

Untersuchen Sie Ihre Anforderungen. Machine Learning kann helfen.
© Fraunhofer IML
Unternehmen, Kunden und Produkte sind verschieden. Daher ist immer zu analysieren, mit welchen Verfahren sich die besten Prognosegüten erzielen lassen. Manchmal liefern konventionelle statistische Verfahren das beste Ergebnis.

Do you want to use the Potentials of Machine Learning? Please talk to us!

We provide a holistic view on supply chains and, what’s more, we also have long-term cross-sectoral experiences in implementation of demand forecast and optimization of supply chains apart from artificial intelligence.