hardDesignCase StudyMidSeniorStaffLead
Vehicle routing is NP-hard. How do you produce a plan the depot will run tomorrow morning?
7 min readUpdated 2026-08-15
Answer
Short answer
You stop pursuing optimality and engineer for a wall-clock budget: build a feasible plan with a construction heuristic, improve it with local search or a metaheuristic until the clock runs out, and model the constraints that genuinely bind, which are time windows, capacity and statutory driver hours.
Was this answer helpful?
Likely follow-ups
- A driver calls in sick at 06:40. What do you run, and how long do you have?
- How would you tell whether your solver or your distance matrix is the cause of a bad plan?
- Two depots could serve the same cluster of stops. How does that change the problem?
- How would you measure whether the plan you produced was any good, after the fact?
Related questions
- Your routing optimiser produces a good plan and the transport planner changes it every morning. What is going wrong?hardAlso on vehicle-routing and optimisation4 min
- How do you decide whether to use a managed service or self-host a component, and which cloud costs catch teams out?hardSame kind of round: design6 min
- Design the account-information API a third party will call on behalf of your customers. Where does consent live, and what does it constrain?hardSame kind of round: design6 min
- Design the gate that decides whether a newly trained model gets promoted.hardSame kind of round: design6 min