06
Simulation and mathematical models for operating and investment decisions
We build a model of your operation and test decisions in it before the money is spent: how many berths, how many wagons, which inventory policy and what a queue does to the system.
Context
A spreadsheet calculates averages. A real operation has queues, breakdowns, rain, late vessels and trucks that arrive all at once. Capacity is lost in that variation, which is why a terminal sized on the average stalls at peak.
Discrete-event simulation reproduces the operation event by event, with the variability measured in the client’s data. Mathematical optimization selects the best configuration among millions of combinations. We use the two separately or together, depending on the question.
The model is validated against history before it is used for decisions. Only then do we test the scenarios: one more berth or one more shift, own or contracted fleet, a new yard or a new service rule. The answer comes in capacity, cost and service level, with the uncertainty range stated.
Scope of work
Simulation model of the operation
A terminal, yard, line, fleet or complete chain reproduced in simulation software, with animation and validation against real data.
Capacity sizing
Number of berths, lines, hoppers, wagons, trucks or storage positions needed to handle the projected volume at the required service level.
Network, mix and schedule optimization
Mathematical programming models for location, flow allocation, sequencing and the scheduling of production or shipments.
Investment scenarios
Comparison of alternatives in works, equipment and operating rules, with the gain from each and the recommended order of execution.
Model handed over and team trained
The client receives the model, the documentation and the training to run new scenarios without depending on us.
Discuss my case
Signals this applies to you
When three or more of these describe your operation, we usually find enough value to justify a diagnostic.
Discuss my case- A capacity investment about to be approved without scenario testing.
- A recurring queue or delay whose cause has not been isolated.
- A new terminal, yard or spur project that has to demonstrate its capacity to a lender.
- A decision between own and contracted fleet, or between more shifts and more equipment.
FAQ
Frequently asked questions
What is the difference between simulation and optimization?
Simulation shows what happens if a decision is taken, considering the variability of the operation. Optimization indicates which decision is best within the constraints. Sizing a terminal calls for simulation. Designing a network or scheduling shipments calls for optimization. In large projects, each feeds the other.
What data is needed?
Arrival and service records, operating times by step, stoppages and volumes by period. Part of the early work is to treat this data and measure the real variability of each step.
Does the client keep the model?
Yes. We deliver the file, the documentation and the training. The annual update is optional.
Does it work for a project that does not exist yet?
Yes. For new projects we use parameters from comparable operations and from the engineering design, and the model then supports the feasibility study and the discussion with the lender.
Related practices
Logistics network optimization and service level
We model the network formally to answer where to produce, where to hold inventory, how to route and at what service level. The solution carries a mathematical guarantee of optimality, which a spreadsheet heuristic cannot give.
02Tax-balanced network design
We optimize the network and the tax structure inside the same model. In Brazilian interstate operations, a decision on where to hold inventory that leaves ICMS out is taken with half the information.
03Operating cost efficiency across the chain
We target the costs that accumulate outside base freight (demurrage, detention, storage, handling and accessorials), with documented recalculation and correction of the root cause.
Next step
Put a number on the recoverable value in your operation
The first step is a 30-minute conversation about your corridor, your contracts and the order of magnitude involved. There is no pitch deck and no commitment.