Plans that go stale overnight
Building a plan takes days. By the time it is ready, demand, capacity or transport has already moved.
Turn complex operations into confident decisions.
OR Genie is a consulting partner for organisations whose plans, schedules and forecasts have outgrown spreadsheets. We model your real constraints, predict what is coming, and build decision systems your teams use every day.
- Optimization & simulation
- Production-grade AI/ML
- Forecasting & analytics
- Marketing mix modelling
Optimal route
Roads
Closed
- Mixed-integer programming
- Constraint programming
- Network optimisation
- Vehicle routing
- Marketing mix modelling
- Discrete-event simulation
- Demand forecasting
- Budget optimisation
- Machine learning
- Generative AI & RAG
- Scenario planning
- Metaheuristics
- Causal experimentation
- MLOps
Most operations already have the data. The decisions are still made by hand.
These are the patterns we hear when a business has outgrown intuition and spreadsheets.
Too many constraints to juggle
Shifts, machines, fleets, service levels and compliance create millions of combinations no planner can search.
Forecasts nobody trusts
Point forecasts without uncertainty push teams to override them, so buffers grow and working capital gets trapped.
Models stuck in notebooks
Promising pilots never reach the planning cycle because nobody engineered them into the tools people use.
Five practices, one goal: better decisions.
We combine mathematical optimisation, machine learning and software engineering. The method follows your problem rather than our preference.
01
Operations Research
Optimisation and simulation for planning, scheduling, routing and allocation. We formulate your objectives and constraints mathematically, then solve them with the method that fits.
02
AI Solutions & Engineering
Production-ready machine learning and generative AI. Models, services and interfaces that run inside your planning cycle, with monitoring built in.
03
Data Science
Forecasting, experimentation and analytics that come with their uncertainty, so plans can account for risk instead of ignoring it.
04
Decision Intelligence
We bring data, models and business context together around a specific decision. Trade-offs are explicit, recommendations are explained, and outcomes feed back into the next decision.
05
Marketing Optimisation
Marketing mix modelling, budget allocation and promotion planning. We measure what each channel really drove, find where extra spend stops paying, and turn it into a budget plan your team can act on.
Meet Geniemix, our MMM product
Built for the industries where constraints are real.
Every industry has its own hard problems. Pick one to see what goes wrong and how we help.
Schedule re-optimised
Changeovers minimised
Manufacturing
Orders, machines, changeovers and maintenance windows all compete for the same hours. Spreadsheet plans break the first time a line goes down.
Where we help
- Production planning and master scheduling
- Shop-floor sequencing that reduces changeovers
- Capacity planning and line balancing
- Predictive maintenance and quality analytics
MILPConstraint programmingMachine learning
Routes re-planned
Time windows met
Supply Chain & Logistics
Every lane, depot and truck trades service against cost. Small routing and stock-placement choices add up to large freight and inventory bills.
Where we help
- Vehicle routing with time windows and capacities
- Network and depot location design
- Load planning and fleet sizing
- Multi-echelon inventory placement
Vehicle routingNetwork flowMetaheuristics
Forecast updated
Replenishment planned
Retail & CPG
Demand moves with seasons, promotions and weather. Too much stock ties up cash. Too little loses the sale and the customer.
Where we help
- SKU and store-level demand forecasting
- Replenishment and allocation across stores
- Price, markdown and promotion optimisation
- Assortment and space planning
ForecastingPrice elasticityStochastic optimisation
Budget re-allocated
Saturation point found
Marketing & Consumer Brands
Every ad platform takes credit for the same sale. Budgets end up following those claims instead of results, and nobody can say which channel still has room to grow.
Where we help
- Marketing mix modelling across online and offline media
- Budget allocation within spend limits and saturation
- Promotion and trade-spend effectiveness
- Incrementality tests that calibrate the model
Bayesian MMMResponse curvesBudget optimisation
Prefer self-serve? See Geniemix
Pick path optimised
Dock slots balanced
E-commerce & Warehousing
Order profiles change every day. Slotting, picking and dock schedules designed for last year slow down this year's peak.
Where we help
- Slotting and storage-location optimisation
- Pick-path, batching and wave planning
- Dock door and labour scheduling
- Fulfilment-node selection for each order
Combinatorial optimisationSimulationForecasting
Recovery plan ready
Crew legality checked
Aviation & Transport
Aircraft, crews, gates and maintenance are tightly linked. One late inbound flight can ripple across the network within hours.
Where we help
- Fleet assignment and aircraft routing
- Crew pairing and rostering
- Gate and stand allocation
- Disruption recovery and schedule resilience
Column generationSet partitioningSimulation
Theatre slots planned
Bed flow simulated
Healthcare
Beds, theatres and clinicians are scarce, and arrivals are uncertain. Queues form wherever capacity and demand don't line up.
Where we help
- Operating-theatre and clinic scheduling
- Bed management and patient-flow planning
- Nurse and clinician rostering
- Simulating capacity changes before making them
Discrete-event simulationRosteringForecasting
Campaign scheduled
Expiry risk flagged
Pharma & Life Sciences
Batch processes, shelf life, quality release and cold chains make every planning decision tighter, and every mistake more expensive.
Where we help
- Campaign and batch scheduling
- Clinical and commercial supply planning
- Expiry-aware inventory policies
- Cold-chain distribution planning
MILPStochastic planningForecasting
Dispatch scheduled
Load forecast updated
Energy & Utilities
Supply has to match a demand curve that keeps moving, while renewables, storage and outages add uncertainty.
Where we help
- Load and renewable-generation forecasting
- Unit commitment and economic dispatch
- Battery and storage scheduling
- Maintenance and outage planning
Time-series MLMILPScenario analysis
Fleet re-dispatched
Blend on spec
Mining & Metals
Material moves through pits, crushers, stockpiles, rail and port. A saving in one step can starve the next.
Where we help
- Truck and shovel dispatch
- Blending to grade and quality specifications
- Mine-to-port logistics planning
- Short-term mine and production planning
Network flowBlending LPSimulation
Cash plan optimised
Risk score explained
Banking & Financial Services
Cash, risk and service capacity have to be planned across thousands of branches, ATMs and customers, in a way regulators and teams can follow.
Where we help
- Cash and ATM replenishment optimisation
- Credit-risk and collections models
- Branch and service-network planning
- Fraud and anomaly detection
Explainable MLOptimisationForecasting
Roster balanced
Interval forecast ready
Contact Centres & Service Operations
Contact volumes swing by the hour. Rosters built on averages leave agents idle in the morning and queues long by the afternoon.
Where we help
- Interval-level volume forecasting
- Agent scheduling and shift design
- Skill-based routing
- Field-service technician routing
Queueing modelsWorkforce optimisationForecasting
The decisions we help you get right.
Filter by practice to see where each kind of work applies. Many problems combine more than one.
Demand forecasting
Forecasts at the level your plan actually uses, with uncertainty included.
Production scheduling
Sequences that respect machines, changeovers and due dates.
Vehicle routing
Routes and loads that meet time windows at lower cost.
Inventory optimisation
Stock levels and reorder policies balanced against service targets.
Network design
Where to place plants, warehouses and lanes, and what each should carry.
Workforce rostering
Shifts that match demand, skills, labour rules and preferences.
Pricing & promotions
Prices and markdowns linked to demand response and supply limits.
Marketing mix modelling
What each channel really drove, separated from sales that would have happened anyway.
Media budget allocation
The channel split that earns the most within your budget, limits and saturation.
Incrementality testing
Geo and holdout tests designed to check what a channel adds, and to calibrate the model.
Simulation & digital twins
Test a new policy, layout or capacity change before you commit to it.
Predictive maintenance
Failure risk turned into maintenance windows the schedule can absorb.
GenAI planning copilots
Assistants that answer questions over your plans, documents and data.
Scenario planning
Compare what-if scenarios side by side, with the trade-offs made explicit.
Risk & anomaly detection
Early warnings that come with a recommended next action, not just an alert.
Dashboards describe. We help you decide.
Analytics shows what happened. Decision intelligence recommends what to do next and shows why.
Typical reporting
- Reports what happened last month
- Leaves the trade-off to gut feel
- Static once it is delivered
- Insight with no clear next action
Decision intelligence
- Recommends what to do next
- Finds the best plan within your constraints
- Re-plans when conditions change
- Explains which constraint drove the answer
Mathematical rigour
Problems are formulated precisely, with clear objectives and constraints, so the answer is defensible rather than approximate.
Business context first
We start from how your operation runs. A constraint the floor can't break is part of the model, not a footnote.
End-to-end engineering
Data pipelines, models, APIs and interfaces. We deliver a system that runs, not a notebook or a slide deck.
Explainable by design
Planners can see why a recommendation was made and what would change if a constraint moved.
The right method, not a favourite
Solver, model or simple rule: the method follows the structure of the problem. If something simpler works, we use it.
Built for adoption
Training, documentation and monitoring are part of the work, so the solution keeps earning its place after handover.
Bring us the decision that is costing you most.
Tell us about the choice, the constraints and what a better answer would change. We'll reply with an honest view of whether optimisation, AI or analytics can help.