This program simulates the operation of an elevator system. The simulation is used to investigate the scheduling algorithm based on different traffic scenarios. There is also a terminal GUI to follow the current progress.
The simulation is modelled as microservices that are deployed for each components:
- controller: schedules the destination of the elevators
- elevator: "workers" that processes the queue at each floor
- floor: maintains a queue of waiting passengers
Below is the flow of messsages sent between the services for sending a passenger to their destination floor.
I have defined a definition in YAML format to describe passengers arriving at a floor at a certain interval and their destinations.
The following scenario definition describes two scenarios:
- 20 passengers arriving at floor 0 and wants to go to floor 0 at time 0
- 10 passengers arriving at floor 1 and wants to go to floor 2 at time 5
- time: 0
passengers:
- start: 0
destinations:
1: 20
- time: 5
passengers:
- start: 1
destinations:
2: 10- Builds the image and run in background:
docker-compose up --build --remove-orphans -d - Start gui:
cd gui; python3 dashboard.py - Send input to floors:
cd input-feeder; python3 input_feeder.py -samples samples/one_at_a_time.yaml
Example simulation running with GUI:
