traffic-demand-tools

Task 1. TOD plan based on smart-intersection data

Sample Code

python TOD.py --crsrd-id 1850041700 --input-dir ./data --output-dir ./result --max-tod 4

Arguments

• crsrd-id : smart-intersection ID
• input_dir : directory including input datasets
• output-dir : directory to save the results
• max-tod : maximum number of TOD groups (> 1)

1-1. Description

Using dataset from smart-intersection, the time table with TOD labels is estimated by K-means method

• Time units: 30 minutes
• Single intersection(crossroad)
• Note: Go-direction traffic includes right-turn traffic TOD considers day types - weekdays, saturday, and sunday

1-2. Requairements

Environment/Packages/Libraries

• Python 3
• pandas
• dplython
• scikit-learn

Input datasets

• ORT_CCTV_5MIN_LOG
• ORT_CCTV_MST

1-3. Results

TOD table

• For each day type, time (30 min-unit), TOD is labeled
• Example

Traffic Analysis

• Traffic characteristics according to each TOD period

• turning rate
• total traffic (veh/30min)

• Example

1-2. Issues

Missing values in input data (ORT_CCTV_5MIN_LOG)

• (1st) moving average
• (2nd) fill with 0 and drop na or inf values during normalization
• Still need to improve

Task 2. RSE Analysis (for target area)

Sample Code

python uniq_rse_analysis.py ./data/RSE_COL_20211124.xlsx 20211124

Arguments

• Target RSE Data File Name
• Target Date (on which the RSE Data is collected)

Description

• Analyze the travel time between specific RSE spots, and draw/store the Boxplot Figure of the travel time according to RSE spots.

• Output Example

Task 3. (Advanced) TOD plan considering SA intersection group

Sample Code

python tod_generator.py --input-dir ./data --output-dir ./result --max-tod 10

Input datasets

• traffic_input.csv
• crsrd_sa.csv

Results

TOD table

• For each SA, day type, time (1 hour-unit), TOD is labeled