READ ME File For 'Analysis of EV Charging Stations Using Game Theory'

ReadMe Author: EFSTATHIOS ZAVVOS, University of Southampton

This dataset supports the publication:
Efstathios Zavvos
'Planning and Analysing Competing En-route Charging Stations 
for Electric Vehicles: A Game-theoretic Approach'
PhD Thesis

This dataset contains the experimental data produced by the author, which support the above mentioned thesis. More detailed explanation follows below.

Date of data collection: Feb 2014 - May 2019

Information about geographic location of data collection: University of Southampton, U.K.

Licence: CC BY-NC

Date that the file was created: June, 2019


DATA FORMATS
------------

Some data are provided in .csv format, and some in .mat format, which require Matlab R2016b or later to open. Figures are in .fig format and require Matlab R2016b or later to open. To load .csv files in Matlab, use the "X = csvread('file.csv')" command to create a matrix from the file, and to load .mat files, the easiest way is to double-click the .mat file from within Matlab, which loads the data to the Workspace. Figures are provided in .fig and .eps format.

INDEXING
--------

There are two types of experiments; short ones that only utilise one stage of the extensive-form game with the other parameters set, and full experiments that utilise two or more stages of the game (i.e. there is a SPE). In all experiments, parameters are kept constant while one parameter is varied. Experiments are organised in Chapters that correspond to the chapters in the thesis, and in the order they are presented. Folder names are descriptive of the parameter that is varied each time, and each experiment also contains a brief description.

HOW TO READ MATRICES
--------------------
Files will be explained in more detail below, but there are a few things that are common across all experiments. First, each experiment involves a given amount of stations and investors. In everything that concerns each individual station e.g. capacity, fee, probability, station utility, driver utility for choosing that station, results are presented in the following way. Let's use capacity as an example. If there are two investors that have 2 locations available each, first is the capacity for investor 1 in location 1, then for investor 1 in location 2, then investor 2 in location 1, and investor 2 in location 2. When there is a train option, probabilities and utilities are then accompanied by a column which is for the train option. So with the above example, after the four probabilities for the stations would be a column for the probability of taking the train. 

	FULL EXPERIMENTS
	----------------
	In full experiments there are 7 .csv files, and a .mat file. Each file will now be 	explained in more detail.
	
	'variables.mat' - This file contains the parameter settings for the experiment. 
	
	'centralised.csv' - Contains the results for a centralised optimum allocation of 		stations. Follows the following format: 
	parameter, capacities, fees, charging 	times, aggregate station utility, 			probabilities, driver utilities, expected utility of the mixed strategy.

	'driver_equilibrium.csv' - Contains data relevant to the drivers' equilibrium.
	Format:
	parameter, probabilities, utilities, expected utility of the mixed strategy

	'expected_queuing_time.csv' - Contains expected queuing times, excluding the time 		needed to recharge driver i. Time is in half-hours (multiply by 30 for minutes)
	Format:
	parameter, expected queuing times for each station

	'runtimes.csv' - Contains runtime information.
	Format:
	parameter, runtime in seconds, average IBR iterations to find SPE

	'settings.csv' - Contains parameter settings for the model

	'station_equilibrium' - Station equilibrium information
	Format:
	parameter, capacities, fees, utilities, charging unit power output (option 1 			represents 50kW, option 2 80kW where applicable)

	'station_investor_utilities.csv' - Contains station and investor utilties
	Format:
	parameter, stations, investors, sum of utilities


	SHORT EXPERIMENTS
	-----------------
	In simple experiments, there are 6 data files.
	
	'capacities.mat' - Station capacities
	'expected_queues.mat' - First are individual station expected queuing times (in 			half-	hours), and then is expected queuing time for driver i.
	'runtime.mat' - contains average runtime for finding an SPE, and avg IBR 			iterations
	'stationU.mat' - Contains station utilities
	'driverU.mat' - Contains driver utilities. First are utilities for each individual 			station, then is the expected utility of the mixed strategy.
	'xvalues.mat' - a vector with the value of the varied parameter each step of the 			experiment.
	
	In full experiments, the first column in each file is always the value of the 			parameter that is varied.
