Transforming internal business processes to deliver predictive maintenance recommendations to customers
Manvelian, Marina (2026)
2026
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202605038926
https://urn.fi/URN:NBN:fi:amk-202605038926
Tiivistelmä
As technological progress continues to accelerate, industrial companies increasingly operate within the data-driven paradigm of Industry 4.0, where Artificial Intelligence (AI) plays a growing role. Maintenance services are one of the areas where emerging technologies can bring significant benefits by shifting from reactive or periodic to predictive maintenance approach. This shift can reduce production costs, improve operational efficiency and support sustainability by enabling early detection of maintenance needs.
The primary objective of this thesis was to investigate how equipment and maintenance service providers can transform their existing non-predictive maintenance processes to support AI-aided predictive maintenance. The thesis applied a Case Study approach, supported by a literature review, and investigation of the case company’s internal processes, and semi-structured interviews with customers and internal stakeholders. The expected outcomes included: a literature-based overview of typical processes for delivering predictive maintenance services; an analysis of the case company’s current maintenance processes; insights into customer expectations; and development proposals for transitioning to AI-aided predictive maintenance.
The focus of this thesis was on service provider’s internal business processes, while an in-depth review of customers’ internal processes and detailed technical solution analysis were outside the scope. The empirical context centered on aggregates customers in Finland.
The results showed that predictive maintenance processes typically include continuous monitoring, anomaly detection, criticality assessment, actionable recommendations and a feedback loop for continuous improvement. The internal process analysis confirmed that integrating predictive capabilities will require, in addition to the technical solution itself, also changes to process logic, roles, resourcing and customer-facing service practices. Two high-level future-state process options were developed to illustrate alternative levels of automation. Customer interviews indicated that predictive maintenance is perceived as valuable when recommendations are accurate, specific, verifiable, timely and easy to act upon. These combined insights enabled the development of a set of recommendations to support organizations transitioning towards predictive maintenance practices.
Future research could examine other customer segments and regions, compare practices and technical solutions across peers and competitors, and explore long-term implications of predictive maintenance for service providers and end customers.
The primary objective of this thesis was to investigate how equipment and maintenance service providers can transform their existing non-predictive maintenance processes to support AI-aided predictive maintenance. The thesis applied a Case Study approach, supported by a literature review, and investigation of the case company’s internal processes, and semi-structured interviews with customers and internal stakeholders. The expected outcomes included: a literature-based overview of typical processes for delivering predictive maintenance services; an analysis of the case company’s current maintenance processes; insights into customer expectations; and development proposals for transitioning to AI-aided predictive maintenance.
The focus of this thesis was on service provider’s internal business processes, while an in-depth review of customers’ internal processes and detailed technical solution analysis were outside the scope. The empirical context centered on aggregates customers in Finland.
The results showed that predictive maintenance processes typically include continuous monitoring, anomaly detection, criticality assessment, actionable recommendations and a feedback loop for continuous improvement. The internal process analysis confirmed that integrating predictive capabilities will require, in addition to the technical solution itself, also changes to process logic, roles, resourcing and customer-facing service practices. Two high-level future-state process options were developed to illustrate alternative levels of automation. Customer interviews indicated that predictive maintenance is perceived as valuable when recommendations are accurate, specific, verifiable, timely and easy to act upon. These combined insights enabled the development of a set of recommendations to support organizations transitioning towards predictive maintenance practices.
Future research could examine other customer segments and regions, compare practices and technical solutions across peers and competitors, and explore long-term implications of predictive maintenance for service providers and end customers.