Modeling bottlenecks in a conveyor roller production line in JaamSim

[WILLIAN SATOSHI YAMAUTI]

Hi everyone,

I'm currently building a JaamSim model for a conveyor roller manufacturing system on my Factory Layout course ate my University in Brazil.

The production process includes:
- Tube cutting
- Machining
- Assembly
- Inspection

Each stage has defined processing times, and machining is intentionally the slowest step, so I expect it to behave as the system bottleneck.

However, I'm not observing a clear queue buildup before the machining stage, and the flow of entities appears more uniform than expected.

What I expected:
A visible bottleneck at machining, with queue accumulation and higher waiting times upstream.

What is happening instead:
- No significant queue forms before machining
- The system appears balanced, even though machining has the longest processing time

What I’ve tried so far:
- Increased machining processing time
- Limited the number of resources at the machining stage
- Checked queue connections and capacities

My current assumptions:
I suspect the issue may be related to one or more of the following:
- The system might not be saturated (arrival rate too low)
- Resource constraints may not be properly enforced
- Lack of variability in processing times (currently mostly deterministic)
- Simulation time may not be long enough to reach steady-state
*According to my AI searches.

My question:
What is the best way to ensure that bottlenecks emerge correctly in JaamSim models?
Are there recommended practices to validate that a resource is truly acting as a bottleneck (e.g., specific metrics or configurations)?

Thanks in advance for any guidance!
Este e-mail foi enviado por um ALUNO da Universidade Federal de Ouro Preto e não reflete a posição da Instituição.