Note
A multi-criteria MILP for intervention planning on ~1,500 national bridges
A mixed-integer linear programme over about 1,500 national bridges that schedules interventions under a multi-period budget and integrates two criteria, instead of prioritising by a single score.
| Organisation | Universidad de los Andes, with INVIAS (Colombia's national roads institute) and Universidad Javeriana |
|---|---|
| Sector | Infrastructure asset management |
| Period | 2020 · B.Sc. thesis and research assistantship (contract VICE 3001) |
| Role | Author of the thesis; research assistant on the INVIAS bridge-management project (2020-2). Advisor: Andrés Medaglia. |
| Stack | Python · gurobipy |
Problem
Colombia's national roads institute (INVIAS) maintains a stock of about 1,500 bridges with a limited annual budget. Prioritising by a single criterion — condition, say — ignores the trade-off with the other criteria that planners actually weigh, and ignores that a bridge postponed today deteriorates and costs more later.
Model
- Decision. Which bridges to intervene in which period, and how.
- Dynamics. Deterioration as arcs between condition states over time, so postponement has an explicit cost.
- Budget. Multi-period, with the budget as the binding constraint.
- Objective. Two criteria integrated in one mixed-integer linear programme, which makes the trade-off between them explicit instead of implicit in a ranking.
- Solver. gurobipy.
Status
Thesis in Spanish (grade 5.0/5.0), with an English summary available on request. Part of the INVIAS–Uniandes–Javeriana agreement on bridge management; the raw bridge inventory is INVIAS data and is not published here.