How to Simulate Intralogistics and Material Flow
To simulate intralogistics, you build a model of how material moves through the plant over time: routes, forklifts and autonomous carts with their speeds, buffers with their capacities, the rules for who calls transport and when, and the order mix that creates the demand. The model then runs a shift or a week and shows where parts wait for pickup, how busy each vehicle is and how full each buffer gets. Those numbers decide whether a line gets its material on time, and a layout drawing cannot show them.
Why material flow problems hide in averages
Most plants size internal transport with a spreadsheet. In an illustrative case, one forklift can move 20 pallets an hour and the line needs 15 pallets an hour, so one forklift looks like enough with room to spare. In this illustrative case the forklift is free a quarter of the time on average.
On the floor, requests do not arrive evenly. Three stations finish a batch within minutes of each other while the forklift is at the far end of the hall. Two vehicles meet in a narrow aisle. The driver's break falls in the busiest hour. So the forklift has spare time on average and is still busy exactly when it is needed.
Averages hide queues, and queues are where intralogistics loses output. A simulation runs every request, every trip and every wait in sequence, so the queues show up before the layout is built or the fleet is ordered. In a stochastic simulation, transport and cycle times also vary from trip to trip, as they do on the floor.
What to put into an intralogistics model
A useful material flow model leaves out most of the detail and keeps the things that create waiting:
- routes: aisles, one-way sections, crossings, dock doors and the distances between stations, racks and buffer areas, usually taken from the CAD layout, which is the base for the model
- vehicles: forklifts, tuggers and autonomous carts, with their number, speed, load capacity and, where it matters, acceleration and braking
- buffers and storage: the capacity of each rack, buffer area and intermediate warehouse, and what happens when one is full
- rules: who calls transport, priorities, batch sizes, milk run frequency and the order in which material is picked
- demand: production orders and product mix for a typical day or week, plus the peaks
- people: drivers and pickers, their shifts and breaks
Most of this data already exists in CAD, the ERP or routing sheets. Missing transport times can be measured with a stopwatch.
What the simulation shows that a spreadsheet does not
A material flow simulation gives numbers you can compare across variants: waiting time for pickup, vehicle utilization, buffer fill level hour by hour and minutes a station waits for material.
| Question | Spreadsheet answer | Simulation answer |
|---|---|---|
| How many forklifts do we need? | Total transport hours divided by hours per shift | Utilization of each vehicle over the shift, including peaks and the queue of open requests |
| Is the buffer big enough? | Average stock level | Fill level hour by hour, and how often it is full or empty |
| Where do parts wait? | Not visible | Waiting time at each station and buffer |
| Is the shortest route the best layout? | Usually assumed yes | Variants compared on output per day, not only on distance |
| What if one vehicle breaks down? | Not visible | Output and waiting time with one vehicle less |
The last row matters more than it looks. A flow that only works when every vehicle runs is fragile. We cover how to test that in How to Compare Resilience, Not Just Throughput.
Real case: where the buffer goes on a new line
A car lamp manufacturer in southern Poland, with 270 people on two shifts and a 28,000 m2 plant that includes a 12,000 m2 warehouse, needed a new production line layout. The plant manager wanted the highest throughput and the best use of labor. The hard part was the time material spends moving between workstations and intermediate warehouses.
DBR77 built a digital twin of the material flow. It used working hours from project timings, standard transport speeds, the acceleration and braking of the trucks, and animation paths. Three layout variants were compared on material flow time and productivity. The result:
- The best layout placed a buffer warehouse in the central area, between production and assembly.
- Its average material flow time was not the shortest of the three.
- Its daily productivity was on average 6% higher than a linear layout without an intermediate warehouse.
- In the other layouts, semi-finished products lost time waiting for pickup between processes.
In other words, the shorter routes in those layouts were eaten up by parts waiting for a vehicle at the end of them. The multi-variant analysis was delivered in about one week. For a method to set up such a comparison, see How to Compare Layout Variants Without Guesswork.
See how material flow is modeled in practice
The use case pages describe projects like the two in this article, from buffer placement on a new line to picking routes in a warehouse.
Real case: picking routes in a warehouse
An HVAC manufacturer in Mazovia, central Poland, with over 150 people on two shifts, makes highly customized products. Its warehouse pick area handles an average of 800 picks per shift with ten people, and the picking was hard to plan.
DBR77 built a digital twin of that warehouse section: transport routes, boundary conditions, operating parameters and transport equipment. Process data was imported in a structured format. An algorithm then generates an optimized picking list adapted to the planned daily product mix. Three scenarios were compared in the simulation:
- Unoptimized, picking in production order sequence: about 1,000 seconds.
- Island algorithm: about 550 seconds.
- Sorting algorithm: about 200 seconds.
The simulation was prepared in a few days from current production orders. Updating it only requires importing revised orders, and the algorithm finds a new solution in seconds, which helps when material or operators are short.
Intralogistics decisions worth simulating
Moving one rack a few meters can be tested on the floor. Simulate when the change is hard to undo or when several things interact:
- Buffer location and size between processes, especially when production and assembly run at different rhythms.
- Fleet size and type: how many forklifts, and whether autonomous carts can take over fixed routes.
- Picking sequence and routes in the warehouse, when the product mix changes every day.
- Milk run frequency and the number of stops per loop.
- Transport during a rebuild, when aisles close for weeks and routes get longer. We cover this in How to Plan Brownfield Changes Without Stopping Production.
A good start is one area where people already complain, such as a line that waits for material.
FAQ
What data do we need to simulate material flow?
A layout with routes and distances, the number and speed of vehicles, buffer capacities and production orders for a typical period. Missing transport times can be measured on the floor. Live machine data, for example from DBR77 IIoT, can be connected later but is not needed to start.
Can the simulation compare forklifts with autonomous carts?
Yes. Each vehicle type is modeled with its own speed, capacity and rules, and the variants are then compared on the same output and utilization figures.
Should the warehouse and the production floor be in one model?
If material moves between them on shared vehicles or through shared buffers, yes. If the warehouse only ships finished goods, a separate model is often enough.
How detailed should the transport model be?
Detailed enough to show waiting. In the car lamp case, the model even included truck acceleration and braking. In a warehouse, routes and picking rules usually matter more.
Conclusion
Start with the area where people already say a line waits for material. Ask the shift leader to log, over a few shifts, each time a station stops because transport did not arrive. If those stops bunch up in certain hours, model that area before you buy vehicles or move racks.
Test your next intralogistics change first
If a forklift count or a buffer size is the open question, the demo shows how routes, vehicles and buffers go into the model and how layout variants are compared on output.
Sources
- DBR77, Case study: Production line design (summary on this site)
- DBR77, Case study: Retrieving material from the warehouse (summary on this site)
- DBR77 Digital Twin, Deterministic vs stochastic simulation
- DBR77 Digital Twin, 2D CAD vs 3D Digital Twin
Want to test a decision from your plant?
Book a demo and we will show how DBR77 Digital Twin compares the options on output, waiting time and transport load.