Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Seeing the Unseen: AI-Driven Discrepancy Detection to Foster Improved Decision-Making and Supply Chain Resilience - A Case Study of PipeChain SCM

Iveberg, Emma LU and Ekstrand, Erik LU (2026) MTTM05 20261
Production Management
Engineering Logistics
Abstract
Background
Supply Chain Management is becoming increasingly more important to organizations in the post Covid-19 era. Managing Supply Chains is also moving towards finding data-oriented ways of working, and companies are making substantial investments in getting digital information systems to integrate with their ways of working. Simultaneously, Artificial Intelligence has become a well-discussed topic over the past few years, with the introduction of large language models and chatbots to the public. There is a massive interest across many industries to find ways in which generative AI can be used to create value, and supply chain companies are no different.
Problem Definition and Purpose
When managing their supply chains, actors often... (More)
Background
Supply Chain Management is becoming increasingly more important to organizations in the post Covid-19 era. Managing Supply Chains is also moving towards finding data-oriented ways of working, and companies are making substantial investments in getting digital information systems to integrate with their ways of working. Simultaneously, Artificial Intelligence has become a well-discussed topic over the past few years, with the introduction of large language models and chatbots to the public. There is a massive interest across many industries to find ways in which generative AI can be used to create value, and supply chain companies are no different.
Problem Definition and Purpose
When managing their supply chains, actors often have access to a number of different systems to monitor operations, e.g. ERP-, customer relationship management- or supplier relationship management-systems. These systems displays data to support system users in their decision making. However, it takes time and effort to reliably be able to learn and navigate from different systems. According to research, one of AI's main strengths is its ability to perform analysis on big data sets. This thesis aims to explore if generative AI can be used to interpret, analyze and present its found insights in supply chain processes to a system user, and thereby making it easier to gather insights from available data.
Method
The thesis deploys a case study, using an abductive research approach. Initially a literature review and a mapping of a case-specific process to establish a foundation of what generative AI will have to interpret and present. Thereafter, current frameworks for AI-prompting are looked into, and then turned into a new one fitting for the case. Finally, the framework is applied to the case, and generated output is evaluated to identify strengths and weaknesses of generative AI's analysis.
Conclusions
The study finds that generative AI can be able to interpret, structure and visualize data in a reliable way in a supply chain setting. The quality and procedure of the analysis is highly dependent on which generative AI model is used. Between the evaluated models, Claude was the best performing, closely followed by ChatGPT, while Gemini and Copilot showed weaker results. Furthermore, all models achieved worse overall output in a test environment where two metrics was to be analyzed compared to only looking at one. (Less)
Popular Abstract
Seeing the Unseen: How AI Can Help Supply Chains Make Better Decisions
When most of us think of artificial intelligence (AI), we either think of chatbots such as ChatGPT, capable of answering questions in a human-like manner, or advanced systems that perform complex analysis of vast amounts of data. But what happens when these capabilities are combined? As digitization has made both historical and real-time data more accessible than ever, a new challenge emerges: determining which information matters and how it should be interpreted. Generative AI may offer a solution.
Imagine driving a car with an abundant amount of warning lights on the dashboard. Some indicate minor issues, while others signal serious problems requiring immediate... (More)
Seeing the Unseen: How AI Can Help Supply Chains Make Better Decisions
When most of us think of artificial intelligence (AI), we either think of chatbots such as ChatGPT, capable of answering questions in a human-like manner, or advanced systems that perform complex analysis of vast amounts of data. But what happens when these capabilities are combined? As digitization has made both historical and real-time data more accessible than ever, a new challenge emerges: determining which information matters and how it should be interpreted. Generative AI may offer a solution.
Imagine driving a car with an abundant amount of warning lights on the dashboard. Some indicate minor issues, while others signal serious problems requiring immediate attention. Now imagine having to determine which lights matter, why they appeared, and what actions to take – all while driving at full speed. This is the reality many supply chain (SC) managers face today. They might open a dashboard containing hundreds of performance metrics, and somewhere within that data, a supplier may be underperforming or inventory shortages may be emerging. The information exists, but finding the most important signals often requires significant time, experience, and manual analysis.
This challenge becomes particularly important in today’s uncertain business environment. Global supply chains are constantly exposed to disruptions caused by geopolitical tensions, changing customer demand, natural disasters, and economic fluctuations. Companies need to identify problems early, understand their causes, and act quickly before small deviations grow into costly disruptions. This is where generative AI could play a new role.
Most people know tools such as ChatGPT for their ability to answer questions and generate text. However, advances in AI have also enabled these systems to analyze structured data and communicate insights in a way that resembles human reasoning. Instead of requiring users to sift through countless reports and dashboards, AI can potentially identify important discrepancies, explain why they matter, and suggest possible actions.
In our thesis, we investigated whether generative AI could be used to support SC decision-making by automatically analyzing KPI data. Working together with a SC software company, we developed a framework that guides AI models through a structured process of identifying deviations, prioritizing the most critical issues, and providing explanations and recommendations.
Several leading AI models were tested using realistic SC scenarios. The results showed that modern AI systems can successfully identify significant discrepancies and generate relevant explanations for why they occur. Rather than simply highlighting numbers outside predefined limits, the AI was able to evaluate the broader context and distinguish between deviations that were minor and those that required immediate attention.
The potential benefits extend beyond saving time. By helping users focus on the most important issues, AI can improve situational awareness and support faster and more informed decisions. In an increasingly volatile world, the ability to recognize emerging risks early may become a crucial capability for building resilient supply chains.
As organizations continue to collect more data than ever before, the challenge is no longer gaining access to information, but understanding it. The future of SC management may therefore belong not to the companies with the most data, but to those that are best at turning data into meaningful decisions.
This popular scientific article is derived from the master thesis:
Seeing the Unseen: AI-Driven Discrepancy Detection to Foster Improved Decision-Making and Supply Chain Resilience – A Case Study of PipeChain SCM, written by Emma Iveberg and Erik Ekstrand. (Less)
Please use this url to cite or link to this publication:
author
Iveberg, Emma LU and Ekstrand, Erik LU
supervisor
organization
course
MTTM05 20261
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Supply Chain Management, Artificial Intelligence, Prompt Engineering, Large Language Model Evaluation, Deviation Detection
other publication id
6064
language
English
id
9241500
date added to LUP
2026-06-25 09:38:12
date last changed
2026-06-25 09:38:12
@misc{9241500,
  abstract     = {{Background
Supply Chain Management is becoming increasingly more important to organizations in the post Covid-19 era. Managing Supply Chains is also moving towards finding data-oriented ways of working, and companies are making substantial investments in getting digital information systems to integrate with their ways of working. Simultaneously, Artificial Intelligence has become a well-discussed topic over the past few years, with the introduction of large language models and chatbots to the public. There is a massive interest across many industries to find ways in which generative AI can be used to create value, and supply chain companies are no different. 
Problem Definition and Purpose
When managing their supply chains, actors often have access to a number of different systems to monitor operations, e.g. ERP-, customer relationship management- or supplier relationship management-systems. These systems displays data to support system users in their decision making. However, it takes time and effort to reliably be able to learn and navigate from different systems. According to research, one of AI's main strengths is its ability to perform analysis on big data sets. This thesis aims to explore if generative AI can be used to interpret, analyze and present its found insights in supply chain processes to a system user, and thereby making it easier to gather insights from available data.
Method
The thesis deploys a case study, using an abductive research approach. Initially a literature review and a mapping of a case-specific process to establish a foundation of what generative AI will have to interpret and present. Thereafter, current frameworks for AI-prompting are looked into, and then turned into a new one fitting for the case. Finally, the framework is applied to the case, and generated output is evaluated to identify strengths and weaknesses of generative AI's analysis. 
Conclusions
The study finds that generative AI can be able to interpret, structure and visualize data in a reliable way in a supply chain setting. The quality and procedure of the analysis is highly dependent on which generative AI model is used. Between the evaluated models, Claude was the best performing, closely followed by ChatGPT, while Gemini and Copilot showed weaker results. Furthermore, all models achieved worse overall output in a test environment where two metrics was to be analyzed compared to only looking at one.}},
  author       = {{Iveberg, Emma and Ekstrand, Erik}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Seeing the Unseen: AI-Driven Discrepancy Detection to Foster Improved Decision-Making and Supply Chain Resilience - A Case Study of PipeChain SCM}},
  year         = {{2026}},
}