Lesson 9 · Batteries and disturbances
A smooth run shows the best a plant can do. Vehicles must also recharge, and in real life things stop for reasons outside the model: something on the floor, a fault, an operator crossing an aisle. You will switch both on and measure what they cost.
- Time
- 30 min
- Start from
- the checkpoint of lesson 8
- You will learn
- enable the battery model and read what it does to the fleet
- add random disturbances, choose their volume and read their cost
- compare runs that include chance, which never repeat exactly
Open your project from lesson 8, or its checkpoint.
The battery model
It is off by default, and optional. When it is on, every vehicle drains its battery as it works, goes to a free charger of its fleet when the charge drops below a threshold, and returns to work once it has recharged to another. The numbers are in two places: the policy (thresholds) is shared by all and lives in the scenario; the pack (type, autonomy, charge time) belongs to each fleet, in the Vehicles tab, where you set it in the last lesson: platforms on lithium, the heavy fleet on lead-acid.
- Switch it on. In Simulate, open the Battery step and tick Simulate charging and discharging. Three fields appear: Initial charge (100 at time zero), Charge below (
30) and Resume from (80). The summary reads the battery types in use and the thresholds, and the charge and discharge curves of each type are drawn below. - Run an hour. Press Start run for
1 h. Nothing changes: with 8 hours of autonomy, nobody gets near the threshold in one hour. The charge under each AGV on the map, and the list in the Results tab, show them slowly draining. - Start low to see the effect. Set Initial charge to
32and run1 hagain. Every vehicle is soon below 30% and, after finishing its transport, drives to a charger and stays there until it reaches 80%. Throughput falls to about half and the queue grows. Open the comparison table to see both runs side by side. - Put it back. Set Initial charge back to
100.
What the experiment teaches: charging takes vehicles out of service, and the slower the charge the more a fleet needs spare vehicles or more chargers. The single heavy truck is the extreme case, because lead-acid takes hours to charge and nobody covers for it. In a real study you would run 8 or 24 hours with a realistic starting charge.
The charge and discharge curves are illustrative, not calibrated on real batteries or chargers. Use them to compare alternatives, not to size a battery.
Random disturbances
The Disturbances step adds stops that come from outside the model. Each is optional and off by default. When you switch one on you choose only its volume, from Low to Extreme on a slider, and the range of seconds a vehicle stays stopped; the duration of each event is drawn at random inside that range. You never choose a rhythm: events strike at random moments, sometimes in clusters, sometimes after long gaps.
| Disturbance | What it does |
|---|---|
| Obstacles on the path | a vehicle that is driving brakes and waits, then resumes by itself |
| Blocking faults | a vehicle at work stops and waits for an operator, then continues |
| Empty pickup / Full drop | a share of pickups find no load, or of drop-offs find the place occupied; with a backup the vehicle tries another station |
| Manual zone reservations | an operator now and then reserves a traffic zone for a manual transport; no vehicle may enter while it is reserved |
- Obstacles and faults. In Disturbances, switch on Obstacles on the path and set its volume to
3of 5 (Medium), then Blocking faults at2(Moderate). Leave the default stop ranges: 5 to 30 s for an obstacle and 120 to 600 s for a fault. - The Repack spur. Further down the list, every traffic zone has its own switch. Switch on
Repack spurat2(Moderate): now and then an operator reserves the spur, and vehicles wait at its edge. An operator cannot reserve a zone with a vehicle inside; that attempt fails and is counted separately. Because the spur starts on the loop, vehicles that wait for it queue on the loop's top side, so the whole loop slows down while the zone is reserved. - Run it twice. Press Start run for
1 hand, when it ends, once more without changing anything. The two runs reproduce the same results because the scenario keeps its random seed. Both rows are in the comparison table, and availability shows the share of time vehicles were not stopped by a disturbance.
Read the Disturbances section of the results: the number of stops, the total time stopped, a count per kind and the last events with their time, AGV and outcome. The KPI strip shows the same at a glance, with the vehicles that are stopped right now.
To study variability, use long runs (8 or 24 hours) with different seeds, as explained in the guide. Use the same set of seeds for each alternative; repeating one seed checks reproducibility, not the spread of possible outcomes.
Stopped vehicles keep their place, so others queue behind them, and a longer stop can expose a deadlock that a smooth run never reached. This is what disturbances are for.
- Both settings on: a battery step with its thresholds, a disturbances step listing obstacles, faults and the Repack zone, and two runs with matching results and the same seed.
- Raise obstacles to Extreme and watch the availability and the queue. At what volume does the plant stop keeping up?
- Switch on Empty pickup with the backup enabled and compare it with the same disturbance without the backup.
- Give the heavy fleet a second truck and a second bay charger and see whether the long lead-acid charge hurts less.
Checkpoint
It contains: 10 paths · 8 stations · 1 traffic zone · 4 AGVs · 2 fleets · 13 workflow blocks.
Download the checkpoint (aurora-09-battery-noise.en.sim)
Open it with Project›Import project. It replaces the project you have open, so save yours first (or open it in a new browser tab). Anything you do from here on is a copy: the checkpoint file itself never changes. Use it to continue from here if you skipped a lesson, or to compare your own project with the expected one.