Examples of flow optimization by simulation

This page brings together several fictitious case studies designed to illustrate different uses of simulation. The objective is not to describe real projects, but to demonstrate simply and visually how these methods can address various industrial challenges.

01 - Determining the size of an AGV fleet before investing (Show)

Determining the right size for an AGV fleet is not simply a matter of adding more vehicles; it is essential to understand the threshold at which the system truly runs smoothly.

This case study demonstrates how simulation can be used to compare different configurations to ensure consistent throughput, avoid bottlenecks, and invest only what is strictly necessary.

An industrialist wants to improve the feeding of his production line.

A question quickly arises: is the current fleet of AGVs sufficient, or should one or more vehicles be added to avoid shortages and maintain the pace?

Without simulation, it is difficult to estimate the real impact of an additional AGV on the entire system.

A model then allows us to test several fleet configurations under the same operating conditions and to compare their effects on flows, expectations and production.

👉 Result: the simulation makes it possible to identify from what level the fleet becomes sufficient to streamline the workshop and improve performance, without overinvesting.

What simulation allows

Simulation allows for an objective comparison of scenarios and the identification of truly effective levers before modifying the organization.

02 - Do we really need to change the bottle-filling machine? (Show)

When a machine is identified as a bottleneck, the first instinct is often to invest in replacing or duplicating it.

However, in this case, the simulation shows that a simple reorganisation centred on ST4 can increase utilisation from 77% to 89%, without modifying the machine.

On a production line, when a workstation is identified as a bottleneck, the first instinct is often to invest: replace the machine, increase its capacity, or add a duplicate.

In this case study, the critical workstation was clearly identified: ST4.

At first glance, the conclusion therefore seemed obvious: if ST4 is limiting the line’s performance, ST4 needs to be upgraded.

But the simulation reveals a more nuanced reality.

In the baseline scenario, ST4 is indeed the most heavily utilised station, but its utilisation rate is only 77%.

In other words, the machine is critical but not yet being utilised to its full potential.

The problem, therefore, stems not only from the machine itself, but also from what is happening around it: workpiece supply, queues, flow coordination, and work organisation.

By testing a reorganisation scenario centred on ST4, without changing the machine, its utilisation rate rises from 77% to 89%.

Output throughput also increases, rising from around 12.2 parts/hour to 14.4 parts/hour.

Same machine.

Different organisation.

Better performance.

This is where simulation becomes a genuine decision-making tool: it allows scenarios to be tested, their impact to be measured, and prevents rushing into a major investment.

Before replacing a machine, one question is worth asking:

Is the current system already being used to its full potential?

03 - Can a production line still be improved when the bottleneck is already operating at 89% utilization? (Show)

Is it possible to improve a production line further when its bottleneck is already operating at 89% capacity?
This case study compares two logistics strategies designed to further optimise the utilisation of the critical bottleneck using simulation.

In a previous simulation model, I worked on optimising the behaviour of an AGV fleet on an assembly line:

reducing unnecessary travel,
improving dispatching rules,
limiting congestion,
and improving workstation feeding.

As a result:
➡️ ST4, identified as the bottleneck of the line, was already reaching an 89% utilization rate.

In other words:
without any major investment, the critical machine was already being heavily utilised.

However, one question remained:

Can performance still be improved without duplicating workstation ST4?

To answer this question, I developed a second, more structural scenario.

This time, the goal was no longer just to optimise AGV movements… but to drastically reduce the need for movement itself.

The scenario relied on integrating a conveyor system in order to:

feed the workstations much more continuously,
reduce transport times,
improve flow fluidity,
and refocus AGVs on more relevant tasks.

Scenario results:

✅ 95% utilization on ST4
✅ 15.7 parts/hour output throughput
✅ a much more stable flow around the bottleneck

And most importantly, an interesting conclusion emerged:

Without duplicating the ST4 machine, achieving significantly higher performance becomes difficult.

At this stage, the simulation shows that the issue is no longer largely logistical.
The bottleneck is now being fed almost continuously.

In other words:
we are probably reaching the limit of this production line architecture.

And this is precisely where simulation becomes a true decision-support tool:

understanding how far an optimisation strategy remains relevant,
measuring the real impact of an architectural change,
and identifying the point at which a heavier investment becomes necessary.

Because in industry, the challenge is not only about optimising a system…

It is also about knowing when the current strategy has reached its limit.

🎥 In the video:
comparison between:

an assembly line optimised through AGV control,
and a line restructured with a conveyor system to maximise bottleneck utilisation.

03 - Autonomous mobile robot (Show)

This case study shows how a dedicated autonomous mobile robot (AMR) for refuelling can relieve operators of trips to the store, reduce unnecessary travel, and increase production, based on a “as is /to be” simulation comparison.

Description

The workshop has 9 workstations, each occupied by 1 operator. To produce, each operator uses parts from boxes stored in the warehouse.

Current model (as is): the operator must go and get a box from the store himself, then return to his station.
Future model (to be): An AMR is responsible for transporting boxes to a dedicated drop-off area in front of each workstation. It anticipates demand and keeps a box on hand in advance, thereby limiting stockouts and back-and-forth trips.

Results (simulation)

Operator travel time: 450 min → 125 min (-72.2%)
Waiting time / non-production: 450 min → 135 min (-70%)
Average distance covered: 14.1 km → 5.4 km (-62%)
Percentage of time spent commuting: 6–21% → 3% (–50% to –85.7%)
Total boxes processed: 112 → 126 (+12.5%)

Conclusion: the integration of a refuelling RMA significantly improves working conditions (less walking, less waiting) and increases productivity, factors that must be weighed against the cost of setting up the robot.

As is model

To be model