Skip to main content

​
Quickstart

Get SimulationBank up and running in minutes. This guide assumes you have Python 3.x and Node.js installed.
For detailed installation instructions, see the Installation Guide.

​
Prerequisites

1

Check Python

python3 --version
# Should show Python 3.7+
2

Check Node.js

node --version
# Should show Node 16+

​
Installation

1

Navigate to backend directory

cd backend
2

Create virtual environment

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
3

Install dependencies

pip install -r requirements.txt
This installs:
  • flask - Web framework
  • flask-cors - CORS support
4

Start the Flask server

python app.py
Server will start at http://localhost:5000

​
Running Your First Simulation

1

Open the application

Navigate to http://localhost:5173 in your browser.
2

Configure simulation parameters

You’ll see configuration sliders for:
  • Number of Tellers: Start with 3
  • Arrival Rate (λ): Try 1.0 customers per second
  • Service Mean (μ): Try 5.0 seconds per customer
  • Simulation Duration: Set to 3600 seconds (1 hour)
For your first run, use default values to see a balanced system.
3

Start the simulation

Click the “Start Simulation” button.You’ll immediately see:
  • Events being processed in the event queue
  • Customers appearing in the priority queue
  • Tellers transitioning between IDLE and BUSY states
  • Metrics updating in real-time
4

Observe the queue visualization

The queue visualizer shows:
  • Red nodes: High-priority customers (Priority 1)
  • Yellow nodes: Medium-priority customers (Priority 2)
  • Green nodes: Low-priority customers (Priority 3)
Notice how high-priority customers are served first, regardless of arrival order.
5

Monitor metrics

The metrics dashboard displays:
  • Wait Time Chart: Average time customers spend waiting
  • Queue Length: Number of customers waiting over time
  • Throughput: Customers served per time unit
  • Saturation Report: Teller utilization percentage
6

Control the simulation

Use the control buttons:
  • Pause: Freeze the simulation at current clock time
  • Resume: Continue from paused state
  • Stop: End simulation and finalize metrics

​
Example Simulation Run

Here’s what a typical simulation looks like: 
# Configuration (backend/src/simulation/domain/simulation_config.py)
config = SimulationConfig(
    num_tellers=3,
    arrival_config={
        "arrival_rate": 1.0,           # 1 customer per second average
        "arrival_dist": "exponential",  # Poisson process
        "priority_weights": [0.1, 0.3, 0.6]  # 10% high, 30% medium, 60% low
    },
    service_config={
        "service_mean": 5.0,            # 5 seconds average service time
        "service_dist": "exponential",
        "service_stddev": 1.0
    },
    max_simulation_time=3600.0,      # 1 hour simulation
    max_queue_capacity=100
)

​
Expected Results

With λ = 1.0 (arrival rate) and μ = 0.2 (service rate = 1/5.0), and c = 3 tellers:
  • Traffic Intensity (ρ): ρ = λ/(c×μ) = 1.0/(3×0.2) = 1.67
  • Since ρ > 1, the system is overloaded - queue will grow unbounded
This configuration will create a growing queue! Try increasing tellers to 6 or reducing arrival_rate to 0.5 for a stable system.

​
Understanding the Output

​
Event Processing

The simulation processes three event types: 
# Customer crosses the door
# - Generate customer with priority and service time
# - Add to waiting queue (sorted by priority)
# - Schedule next arrival
# - Try to assign idle teller

​
Queue Dynamics

Customers in the waiting_queue are sorted by:
  1. Priority (ascending): 1 (High) < 2 (Medium) < 3 (Low)
  2. Arrival time (ascending): FIFO within same priority
# From simulation.py:118
self.waiting_queue.sort(key=lambda c: (c.priority, c.arrival_time))

​
Next Steps

Configure Parameters

Learn how to tune simulation parameters

Interpret Metrics

Understand wait times, throughput, and saturation

Simulation Engine

Deep dive into the DiscreteEventSimulation class

Advanced Scenarios

Explore high-load and optimization scenarios

​
Troubleshooting

Kill the process using port 5000:
# Linux/Mac
lsof -ti:5000 | xargs kill -9

# Windows
netstat -ano | findstr :5000
taskkill /PID <PID> /F
Verify:
  1. Backend is running at http://localhost:5000
  2. CORS is enabled (flask-cors installed)
  3. Check browser console for CORS errors
Your arrival rate is too high for the number of tellers!
  • Increase num_tellers
  • Decrease arrival_rate
  • Decrease service_mean (faster service)
Rule: Ensure ρ = λ/(c×μ) < 1 for system stability
Check your arrival_rate:
  • Very low rates (< 0.1) mean long intervals between customers
  • Increase to 1.0 or higher for visible activity
For a balanced system where queues remain manageable, aim for ρ ≈ 0.7-0.8 (70-80% teller utilization).

Build docs developers (and LLMs) love

Get started for freeTalk to us