“Planning still happens in a spreadsheet and a head”
The best planner knows every exception; the plan is only as good as their day — and they are on holiday in August.
Mentis · Services · Optimization & operations AI
Optimization that moves the P&L — routes, schedules, resources
Route, scheduling, load and resource optimization modelled on your real constraints and your messy data — engines that cut kilometres, overtime and idle capacity in measurable euros, integrated into the systems your planners already use.
When to call us
Sound familiar?
“The off-the-shelf optimizer doesn’t fit our rules”
Time windows, skills, vehicle types, union rules, customer priorities — generic tools make you bend the business to the software.
“We suspect there are millions in small inefficiencies”
Empty kilometres, overtime, overstock, missed SLAs — visible in aggregate, invisible in any single decision.
What you get
Outcomes, not deliverables first
Euros, not dashboards
Savings expressed in kilometres, hours, vehicles and service levels — measured against your current plan before and after, on your data.
Your constraints, encoded
Time windows, skills, capacities, regulations and priorities modelled explicitly — the plan respects reality instead of approximating it.
Planners in control
Decision-support first: the engine proposes, people override, and every override teaches the model — adoption follows.
What we deliver
The scope you can actually sign
Optimization opportunity scan
Two to three weeks on your historical plans and orders: where the slack is, what an optimizer would realistically recover, what it would cost.
Constraint & objective modelling
The business rules, priorities and trade-offs written down and validated with planners — the part that makes or breaks adoption.
Optimization engine
Routing, scheduling, load-building, shift or resource allocation — built with proven solvers and heuristics, benchmarked on your data.
Forecast & demand inputs
Demand forecasting and ETA prediction feeding the optimizer, so plans are built on what will happen rather than what happened.
Integration into planning workflows
APIs and interfaces into your ERP, WMS, TMS or field-service tools; plans land where dispatchers and planners already work.
Monitoring & continuous tuning
Plan-quality dashboards, constraint drift alerts, re-tuning as the business changes — the savings stay real after launch.
How it runs
Fixed scope, fixed price per phase — you decide at each step.
- 1
Opportunity scan (2–3 weeks)
Historical data in, quantified savings potential and a fixed-price proposal out.
- 2
Pilot on one flow (4–8 weeks)
One depot, one region or one team: engine built, results measured against the current plan.
- 3
Roll-out & integrate
Remaining flows, integrations, planner training, monitoring.
Regulation & sovereignty
Lighter regulation — with two exceptions worth designing for
Operational optimization is mostly minimal-risk under the AI Act. The exceptions: systems that allocate work, shifts or performance to people fall under worker-facing obligations and social-partner sensitivities, and driver or employee data stays personal data under GDPR. We design the human-in-the-loop and the data minimisation in from the start — and we deploy on your infrastructure.
Questions
Is this the same as OptimizeFlow?
OptimizeFlow is our product for route and field-service optimization — the fastest path when your problem fits it. This service covers the wider space: custom engines for scheduling, load-building and resource allocation, the forecasting that feeds them, and the integration work, with OptimizeFlow or with your own stack.
How much can we realistically save?
Typical ranges in logistics are 5–15 % of kilometres or hours, more where planning is manual — but we refuse to quote your number before the opportunity scan, which measures it on your data.
Our data is messy — does that kill it?
Messy is normal. The opportunity scan tells us what is usable, what needs cleaning and what can be estimated; optimizers are surprisingly tolerant when the constraints are right.
Will our planners accept it?
Only if they stay in control — which is why we build decision support first, keep overrides, and involve planners in the constraint modelling. Adoption is designed, not hoped for.
Other services
Start here
AI Readiness Assessment
A 4-week, fixed-price diagnostic across six dimensions: maturity radar, scored opportunity portfolio, AI Act exposure register and a 90-day plan — the entry point for most clients.
Learn more →ServiceEU AI Act compliance & governance
Know exactly what the EU AI Act asks of you — and be able to prove it
Learn more →ServiceGenerative AI services
Generative AI that answers from your documents — inside your walls
Learn more →ServiceMachine learning & predictive modelling
Predictions you can explain, defend and act on
Learn more →ServiceSovereign AI architecture & MLOps
AI infrastructure your data never has to leave
Learn more →ServiceCustom AI application development
From a model that works to a product your people actually use
Learn more →Next step
Start with a 30-minute conversation
Tell us the problem, not a spec. You get an honest read on feasibility, data, compliance exposure and a first step — within one business day.