navs-svan / VISSIM-Pedestrian-Calibration

Calibration of the significant Social Force Parameters in Vissim

Vissim Pedestrian Calibration

This repository contains the python scripts of my undergraduate research study. They involve the use of the Vissim COM interface.

• Calibration.py - script to calibrate the significant pedestrian walking behavior parameters under the adapted Social Force Model in Vissim. Calibration was done using Genetic Algorithm.
• Validation.py - script to validate the potential calibrated parameters. Validation was done through Theil's Indicator U and Root Mean Square Percentage Error (RMSPE).
• Analysis_GreenTiming.py - script used to establish an empirical relationship between pedestrian volume and minimum pedestrian green cycle time. Analysis was done by using the calibrated parameters in Vissim.
• Analysis_SpeedDensity - script used to establish an empirical relationship between pedestrian density and walking speed. Analysis was done by using the calibrated parameters in Vissim.

DISCLAIMER: Data collection is not included in this repository.

Prerequisites

1. Python 3

2. Vissim License (Trial version does not have access to COM Interface)

Configuration

• Python

pip install -r requirements.txt