Afstudeerstage AI-Powered Function Point Sizing

Accurate software sizing using Function Point Analysis or Easy Functional Sizing remains a manual, expert-driven process. This thesis explores how Large Language Models can automate this task by translating common project artefacts into traceable size estimates. You’ll investigate which artefacts are essential, assess LLM accuracy versus expert judgment, and design an end-to-end framework that links early project bids to delivery metrics. Today, accurate Function Point Analysis (FPA) or Easy Functional Sizing (EFS) counts rely on experts interpreting heterogeneous artefacts (user stories, APIs, diagrams, code) and aligning their understanding. It is unclear which minimal set and quality of artefacts are required for reliable counting, and how consistently a Large Language Model (LLM) can translate those artefacts into FPA/EFS elements. Additionally, there is no standard workflow connecting early bidding to realised throughput in Function Points. This thesis addresses these gaps.

The Assignment

Investigate how to automate and operationalise FPA/EFS within Info Support’s Way of Working using AI. Deliver a Proof of Concept (PoC) that ingests typical project artefacts and produces traceable FPA/EFS counts, along with a framework to integrate this process from bid to delivery metrics. Evaluate accuracy, consistency and lead time against a human baseline on multiple cases.

• Documentation readiness for FPA/EFS – Analyse which artefacts (and quality criteria) must be present pre‑bid and per delivery phase for accurate counting (e.g., backlog items, acceptance criteria, OpenAPI/AsyncAPI, entity models, sequence diagrams, code). Produce a readiness checklist and examples.

• LLM‑based translation to FPA/EFS – Design and implement a pipeline that maps artefacts to FPA/EFS elements (EI, EO, EQ, ILF, EIF) and/or an EFS estimate with explanations and traceability (why each element was counted). Compare prompting vs. structured extraction, few‑shot exemplars, and rule‑assisted post‑processing.

• Integration framework in the Way of Working – Define how this fits Info Support’s standard process: from bid (initial sizing and pricing), through project start (baseline), to tracking (FP delivered per sprint/context) and retro (variance analysis). Provide reference integrations (e.g., Azure DevOps/Jira for backlog; OpenAPI; dashboards).

Example Research questions

• What is the minimal viable artefact set (and quality level) needed for accurate and reproducible FPA/EFS counts?

• How accurate and consistent are LLM-assisted counts compared to expert counts across domains and artefact types?

• How can we operationalise FP metrics in our Way of Working to improve bidding and delivery predictability?

About Info Support Info Support specializes in custom software, data/AI solutions, management, and training and is active in the Finance, Industry, Agriculture, Food & Retail, Mobility & Public, and Healthcare sectors. We provide solid and innovative solutions for complex and critical software issues. Our headquarters are located in Veenendaal (NL) and Mechelen (BE). At present, approximately 500 employees are employed by Info Support.

Info Support’s working method is characterized by a number of core values: solidity, integrity, craftsmanship, and passion. These core values are intertwined in our work and the way we interact with each other.

To ensure that all employees are always up to date with the latest developments, Info Support has an in-house IT Academy that eagerly satisfies the hunger for more or different knowledge and skills.

B2 language proficiency in Dutch is required.