Shared Parking Calculator

This repository contains a calculator that forecasts shared parking demand for mixed-use sites. Parking demand for each land use is calculated on an hourly basis from 6:00 - 0:00 within a typical weekday or weekend day for each month of the year. The calculator also plots the following graphs:

• Maximum hourly parking demand for each month, line graph with 85th percentile demand
• Hourly parking demand for busiest month, stacked bar graph
• Hourly parking demand for 85th percentile month, stacked bar graph

Workflow Example

The above flowchart and accompanying instrucitons are for an example mixed-use site with retail, food services, entertainment, fitness, office, hotel, and child care uses. Base parking demand is calculated using appropriate parking generation ratios (bylaw, demand, or supply) with respect to variables such as gross floor area, number of hotel rooms, or number of children in care.

1. Categorise individual land uses (cyan) into land use classes (navy blue) with common parking behaviours. For example, Gym and Yoga Studio have similar patron parking patterns so they are modeled as a Health Club. Create as many land use classes as required to capture different parking behaviours, ensure that name and factors defined as per below instructions. Unused land use classes may be retained in the input files.

The following steps refer to files in the Inputs folder

2. Compile the base parking demand for each land use class into the BaseParkingDemand.csv file, typically analyze weekday and weekend scenarios because certain uses such as office are not as active on weekends while shopping and restaurant uses might be busier.

3. Decide on the staff/visitor split as staff and visitors (employees and customers) are suject to separate adjustment factors. The staff/visitor splits are stored in the CustomerEmployeeSplit.csv file.

4. Decide on the time of day factors for each land use class, there will be one set of hourly factors (6:00 - 0:00) for employees and one set for customers for each class. The time of day adjustment factors are stored in the TimeOfDayWeekday.csv and TimeofDayWeekend.csv files.

5. Decide on the non-captive market adjustment factors for each land use class, there will be one set of hourly factors (6:00 - 0:00) for employees and one set for customers for each class. The time of day adjustment factors are stored in the NoncaptiveAdjustmentWeekday.csv and NoncaptiveAdjustmentWeekend.csv files.

6. Decide on the monthly adjustment factors for each land use class, there will be one set of monthly factors for each class, common to both employees and customers. The monthly adjustment factors are stored in the MonthlyAdjustment.csv file.

Values for the above input files are usually based on industry guidance and/or engineering judgement, make sure to document your assumptions.

Once the input file values have been finalised, call the program from the terminal python calculate_shared_parking.py --dir [insert filepath here] where the filepath is where the Inputs and Outputs folders are stored. Once called, the program will perform the following steps:

1. Read base parking demand for each land use class

2. Split the base parking demand into customer parking demand and employee parking demand which are subject to their own set of adjustments.

3. Multiply the base parking demand by the time of day factors to generate an hourly demand vector for each land use. Each land use class will calculate one hourly demand vector for customers and one hourly demand vector for employees to arrive at unadjusted shared parking demand.

4. Apply non-captive market adjustments for each land use, varying by the hour, to account for the interactions between land uses within the site. For example, an office worker going to a fast-food restaurant for lunch on the site is already counted as being parked for the day and would not generate addditional parking demand for the fast-food land use.

5. Apply monthly adjustments for each land use to generate parking demand profiles from January to December.

6. Add the adjusted employee and customer hourly parking demands for each land use class together. The hourly parking demand for the weekday and weekend scenarios are generated in the Output folder and are named WeekdayParking.xslx and WeekendParking.xlsx respectively.

Dependencies

• All code is written in Python 3
• Numpy
• Pandas

File Descriptions

Python files:

filename	description
calculate_shared_parking.py	Calls functions to calculate weekday and weekend parking demand
get_inputs.py	Two functions, one to get directory containing input and output folders from user terminal input and one function to read the input files
LandUse.py	Contains LandUse Class and associated calculations, dataframe manipulations, and output formatting

Input files:

filename	description
BaseParkingDemand.csv	Unadjusted parking demand for each land use class
CustomerEmployeeSplit.csv	Proportions of customers to employees for each land use class, weekday and weekend values
MonthlyAdjustment.csv	Adjustment factors for Jan to Dec for each land use class, one set of factors for both customers and employees, no weekday/weekend distinction
NoncaptiveAdjustmentWeekday.csv	Adjustment factors for non-captive market effects for each land use class, one set of factors each for customers and employees, weekday values only
NoncaptiveAdjustmentWeekend.csv	Adjustment factors for non-captive market effects for each land use class, one set of factors each for customers and employees, weekend values only
TimeOfDayWeekday.csv	Adjustment factors for temporal demand variation for each land use class, one set of factors each for customers and employees, weekday values only
TimeOfDayWeekend.csv	Adjustment factors for temporal demand variation for each land use class, one set of factors each for customers and employees, weekend values only

Output files:

filename	description
WeekdayParking.xlsx	Spreadsheet grouped by month, contains weekday hourly parking demand broken down by land use class, contains total column
WeekdendParking.xlsx	Spreadsheet grouped by month, contains weekend hourly parking demand broken down by land use class, contains total column

Credits

Author: Joshua Cayanan

The methodology is adapted from Shared Parking (Third Edition) by Mary Smith, published by the Urban Land Institute.

Car Parking icon made by monkik from www.flaticon.com
Calculator icon made by srip from www.flaticon.com

To-Do List

• Complete description of files section in README
• Add flowchart of calculation process
• Add feature that plots maximum total parking demand per month as a line graph
• Add feature that identifies 85th percentile total parking demand
• Add graphing feature that plots stacked bar chart of the 85th percentile month and the busiest month
• Add feature that asks the user for the current working directory
• Add GUI

License

This project is licensed under the terms of the MIT License.