Introduction

Thiele GmbH & Co. KG, a leading manufacturer specialising in chains and conveyor technology, operates in industries where equipment reliability is critical, including mining and bulk materials handling. Recurring equipment failures at a customer’s facility highlighted an opportunity to explore predictive maintenance as a means of improving operational reliability and developing new service-based business models.

As part of my Executive MBA at IMD, I conducted a strategic assessment examining the potential application of machine learning to predict equipment failures. Drawing on direct access to Thiele, customer interviews, technical discussions and financial modelling, the project explored how emerging technologies could create additional customer value and new commercial opportunities.

Strategic Overview

Exploring Machine Learning for Predictive Maintenance

The project was initiated following consecutive equipment failures at a key customer’s facility. These incidents highlighted the limitations of reactive maintenance approaches and provided the starting point for investigating data-driven alternatives.

My work involved analysing operational challenges, gathering customer insights and evaluating how machine learning could potentially identify failure patterns and support more proactive maintenance decisions.

Technical Feasibility

Evaluating Machine Learning Applications

The proposed solution explored the potential use of supervised and unsupervised machine learning algorithms to analyse operational data and identify patterns associated with equipment deterioration.

Through discussions with Thiele’s engineering team and technology specialists, including IconPro and Talpa Solutions, I examined technical feasibility, potential data requirements and the commercial implications of a predictive maintenance offering.

Rather than implementing a technical solution, the assignment focused on developing a strategic concept and evaluating its potential business value.

Outcomes and Business Impact

Exploring the Commercial Potential of Predictive Maintenance

The project resulted in a strategic proposal and financial business case exploring the potential application of machine learning to predictive maintenance in industrial conveyor systems.

Based on hypothetical assumptions, the financial model illustrated the potential economic attractiveness of the proposed solution. These projections were developed for academic assessment and do not represent actual implementation results or validated company forecasts.

Beyond financial modelling, the project demonstrated how customer insights, industrial expertise, emerging technologies and business model innovation can be brought together to identify potential growth opportunities.

The experience reinforced my interest in translating technological possibilities into commercially viable strategies while recognising the organisational and operational challenges involved in implementation.

Future Directions

Opportunities for Digital Service Innovation

The proposed concept illustrated how predictive maintenance could potentially extend beyond individual conveyor applications towards broader digital service offerings.

Further development would require validating available operational data, assessing the predictive capabilities of machine learning models and testing the economic assumptions under real-world conditions.

The project highlighted the opportunities and challenges associated with transforming established industrial products into digitally enabled solutions and service-based business models.

IMD Executive MBA Strategic Assignment | 2023

References

Ganeriwalla, A., Harnathka, S., Costa, A., Volkov, M., & Voigt, N. (2021). Racing toward a digital future in metals and mining. Boston Consulting Group. Available at: https://web-assets.bcg.com/65/43/b63e686f453d82fb46f6a0702e2d/bcg-racing-toward-a-digital-future-in-metals-and-mining-feb-2021.pdf [Accessed 16 April 2023].