Data
Read-only queries on the company’s order records. The live system is never written to.
- Several years of order lines
- Made-to-order lines, linked to a production order
- Cancelled, service and test lines left out
Proposal defended, November 2026. Analysis in progress.
Thesis title Using Machine Learning and Agentic AI to Measure, Explain and Predict Order-to-Delivery Lead Time in a Configure-to-Order Manufacturer
In plain words: find where the order time goes, then warn planners before an order runs late.
There are no thesis results yet. The demo runs the same method on synthetic data.
A customer who orders a built-to-spec product cares about two things. Will it arrive on the date they were given, and how long will they wait? That wait runs from the order date to delivery, through sales, engineering, planning, production, inspection and dispatch. The case is a configure-to-order manufacturer: it builds each product from options the customer picks. It quotes lead time from a standard table that leaves out part of the wait. Nobody measures the full wait, and nobody can tell at order entry if an order will be late. So the study puts reliability first and speed second. A shorter quote does not help if it is then missed.
Engineering and approval time the quote leaves out
Split the wait into stages. Find the few stages that carry most of its length, spread and delay.
Check how often the first promised date is met. Compare the promised interval with the wait the customer goes through.
Predict the delivery date and the late risk on the day the order is entered. Compare with today’s promise. Use it to warn early and to target fixes.
Four steps. Each one feeds the next. RQ1 to RQ3 are the three research questions above.
A single case, studied with numbers. Everything rests on the company’s order records, at the level of the order line.
Read-only queries on the company’s order records. The live system is never written to.
Medians and percentiles carry the analysis, because a few orders wait far longer than the rest.
A decision tree first, because planners can read its rules and push back. A random forest shows what that readability costs.
A small Python agent scores open orders each morning and notes any stage held longer than usual. A language model writes the alert in plain words. It has no write access. It flags; people act.
A simulation fed by the stage times tests how on-time delivery and lead time respond when a key stage is cut or steadied.
| Stage | From | To | Years |
|---|---|---|---|
| Pre-production | Order date | Production start | All years |
| Production | Production start | Last unit completed | All years |
| Post-production | Last unit completed | Line fully delivered | All years |
| Order handling and engineering | Order date | Release to planning | Recent years |
| Planning | Release to planning | Production order created | Recent years |
| Production queue | Production order created | Production start | Recent years |
The three finer pre-production stages can be traced only for recent years.
From proposal to submission.
Proposal defended, November 2026. Analysis in progress.
Literature review finished. Proposal submitted and defended.
Proposal defendedCompany approval letter. Pull and clean the order data.
Clean datasetStages, spread and Pareto. The promise check.
RQ1 and RQ2 resultsA paper on RQ1 and RQ2.
Paper submittedRQ3: the decision tree and the random forest.
RQ3 modelsAgent build and replay test. What-if scenarios.
Prototype and demoResults, discussion and conclusions. Submit the thesis. Start a journal paper.
Thesis and journal draftDates may shift with the programme calendar.
A fictional company with about 6,000 synthetic order lines. Time the stages, enter an order and watch the tree score it, replay a year of early warnings, and try two what-if levers.
Synthetic data for a fictional company. These are not thesis results.Open the demo