SimulationConfig

Overview

The SimulationConfig object defines all parameters needed to configure and launch a bank simulation. It controls the number of resources, arrival patterns, service characteristics, and operational limits.

Schema

num_tellers

integer

default:3

Number of teller windows available to serve customers simultaneously.Constraints:

Must be a positive integer
Recommended range: 1-10 for optimal performance

arrival_config

object

Configuration object defining customer arrival patterns.

Show properties

arrival_rate

float

Lambda (λ) parameter for the arrival rate, representing customers per time unit.Example: 1.5 means an average of 1.5 customers arrive per second.

arrival_dist

string

default:"exponential"

Probability distribution model for arrival intervals.Supported values:

exponential - Poisson arrival process (memoryless property)

priority_weights

array

Probability weights for assigning priority levels to arriving customers.Format: [HIGH, MEDIUM, LOW]Constraints:

Must be an array of exactly 3 floats
Values must sum to 1.0
Each value must be between 0.0 and 1.0

Example: [0.15, 0.35, 0.5] means:

15% HIGH priority (elderly, pregnant, disabled)
35% MEDIUM priority (preferred customers)
50% LOW priority (regular customers)

service_config

object

Configuration object defining service time characteristics.

Show properties

service_mean

float

Mean (μ) service time in seconds required to process a customer transaction.Example: 4.0 means average service time is 4 seconds.

service_dist

string

default:"exponential"

Probability distribution model for service durations.Supported values:

exponential - Exponential service time distribution

service_stddev

float

Standard deviation (σ) for service time distribution.Note: Used when distribution requires variance parameter. For exponential distribution, this may be ignored as variance is determined by the mean.

max_simulation_time

float

default:28800

Maximum simulation time horizon in seconds. Simulation will terminate when this time is reached.Default: 28800 seconds (8 hours)Common values:

3600 - 1 hour
14400 - 4 hours
28800 - 8 hours (full business day)
43200 - 12 hours

max_queue_capacity

integer

default:100

Maximum number of customers that can wait in queue simultaneously. Customers arriving when the queue is at capacity will be rejected.Constraints:

Must be a positive integer
Recommended: Set based on physical space constraints

Note: When queue reaches capacity, new arrivals are rejected and counted in the customers_rejected metric.

Example Configuration

Basic Configuration

{
  "num_tellers": 3,
  "arrival_config": {
    "arrival_rate": 1.0,
    "arrival_dist": "exponential",
    "priority_weights": [0.1, 0.3, 0.6]
  },
  "service_config": {
    "service_mean": 5.0,
    "service_dist": "exponential",
    "service_stddev": 1.0
  },
  "max_simulation_time": 28800,
  "max_queue_capacity": 100
}

High-Traffic Scenario

{
  "num_tellers": 5,
  "arrival_config": {
    "arrival_rate": 2.5,
    "arrival_dist": "exponential",
    "priority_weights": [0.2, 0.3, 0.5]
  },
  "service_config": {
    "service_mean": 3.5,
    "service_dist": "exponential",
    "service_stddev": 0.8
  },
  "max_simulation_time": 14400,
  "max_queue_capacity": 150
}

Low-Traffic Scenario

{
  "num_tellers": 2,
  "arrival_config": {
    "arrival_rate": 0.5,
    "arrival_dist": "exponential",
    "priority_weights": [0.1, 0.2, 0.7]
  },
  "service_config": {
    "service_mean": 6.0,
    "service_dist": "exponential",
    "service_stddev": 1.5
  },
  "max_simulation_time": 28800,
  "max_queue_capacity": 50
}

Validation Rules

All configuration parameters are validated on simulation start. Invalid configurations will return a 400 Bad Request error.

Common validation errors:

priority_weights must sum to 1.0
num_tellers must be at least 1
arrival_rate and service_mean must be positive
max_simulation_time must be greater than 0
max_queue_capacity must be a positive integer

Performance Considerations

System Utilization: The ratio of arrival rate to service capacity (ρ = λ/(μ×c)) should typically be less than 1.0 to avoid unbounded queue growth, where:

λ = arrival_rate
μ = 1/service_mean
c = num_tellers

Recommended Configurations

Scenario	Tellers	Arrival Rate	Service Mean	Utilization
Low	2	0.5	6.0	~45%
Medium	3	1.0	5.0	~60%
High	5	2.5	3.5	~87%
Stress	3	2.0	2.5	~97%

Endpoints

Models

Overview

Schema

Example Configuration

Basic Configuration

High-Traffic Scenario

Low-Traffic Scenario

Validation Rules

Performance Considerations

Recommended Configurations

Build docs developers (and LLMs) love

Endpoints

Models

​Overview

​Schema

​Example Configuration

​Basic Configuration

​High-Traffic Scenario

​Low-Traffic Scenario

​Validation Rules

​Performance Considerations

​Recommended Configurations

Build docs developers (and LLMs) love

Overview

Schema

Example Configuration

Basic Configuration

High-Traffic Scenario

Low-Traffic Scenario

Validation Rules

Performance Considerations

Recommended Configurations