Output Data

Request Statistics

tbc

Vehicle Trajectory Statistics

tbc

Standard Evaluations

In this section, we provide an overview of the output file parameters.

General notes about units:

• Time: Measured in seconds from midnight, where 0 represents 12 am (midnight), 16871.94 is approximately 4:40 am, and 36600.32 corresponds to around 10:07 am.
• Distance: Measured in meters, unless specified otherwise.
• Fare: Expressed in cents, with 100 cents equaling 1 Euro.

Output file on user statistics

In the following table, parameters of the file 1_user_stats.csv are described.

parameter	data type	description
request_id	If request is for passenger: int, if request if for parcel: string, starting with “p_”	The unique ID of the request
rq_type	str	The name of the request module based on in the simulation: - BasicRequest (passenger request) - BasicParcelRequest (parcel request)
number_passenger	int	The number of passengers (set in the input request file)
rq_time	int	Time of the request in seconds
earliest_pickup_time	int	Constrain for the earliest pick-up time. Is only filled if different to rq_time, which is the case if in the passenger demand file an extra column is filled for this, see 2_input_data/trips_X.csv
start	str	Network position representation of the start location. For example, 959;-1;-1 means that the start node is 959 and the starting position is exactly at this node.
end	str	Network position representation of end location, see “start”
pickup_location	str	The pick-up location. Generally, this field is not filled as it does not differ from “start”. In order for it to be filled, one had to create a fleet control module.
dropoff_location	str	The drop-off location. Analogous logic as pickup_location
access_time	float	The time from start to pickup_location. This is only filled if pickup_location is filled
egress_time	float	The time from end to dropoff_location. This is only filled if dropoff_location is filled
direct_route_travel_time	float	Fastest travel time from start to end at request_time (in seconds)
direct_route_distance	float	Distance of the fastest route from start to end at request_time (in meters)
offers	str	A list of offers that have been made. For example, if only one offer has been made: 0:t_wait:12.3;t_drive:98.2;fare:100 Operator 0 offers to pickup the costumer in 12.3 seconds, drive her to the destination in 98.2 seconds and demand 1 Euro for it.
decision_time	float	Time that a request accepts an offer (request time if not specified)
chosen_operator_id	id	TODO Only filled if request has been served
operator_id	id	Operator id that served the request Only filled if request has been served
vehicle_id	id	The id of the vehicle that executed the request Only filled if request has been served
pickup_time	float	Time of the actual pick-up in seconds Only filled if request has been served
dropoff_time	float	Time of the actual drop-off in seconds Only filled if request has been served
fare	integer	Fare in cents Only filled if request has been served
modal_state	integer	0 if monomodal trip, else TODO

Output file for general evaluations

The parameters in the file standard_eval.csv are described in the following table:

parameter	data type	description
operator_id	int	Unique ID of the operator
number users	int	Number of bookings made with the operator
number travelers	int	Number of travellers. This can be different from the number of users if there were group bookings (i.e. more than one person is on a booking)
modal split	float	Fraction of travellers that have been served by the operator
modal split rq	float	Fraction of bookings that have been served by the operator
reservation users	int	Number of pre-booked bookings
reservation pax	int	Number of pre-booked travellers
served reservation users [%]	float	Fraction of those pre-booked bookings, that the operator served
served reservation pax [%]	float	Fraction of those pre-booked traveller bookings, that the operator served
online users	int	Number of on-demand bookings
online pax	int	Number of on-demand travellers
served online users [%]	float	Fraction of those on-demand bookings, that the operator served
served online pax [%]	float	Fraction of those on-demand travellers, that the operator served
% created offers	float	Fraction of bookings that have received an offer by the operator. This is different to the parameter modal_split, since passengers can also reject the offer by the operator.
utility	float	TODO (outdated)
travel time	float	Average travel time for all bookings served by this operator in seconds ($ \frac{1}{number_users} \sum_{i \in R_{served}} (\text{drop_off_time}_i - \text{pick_up_time}_i)$ )
travel distance	float	Average distance for all bookings served by this operator in seconds TODO (Currently not computed)
waiting time	float	Average waiting time (difference between request and pickup time) in seconds ($ \frac{1}{number_users} \sum_{i \in R_{served}} (\text{pick_up_time}_i - \text{rq_time}_i)$ )
waiting time (median)	float	Median waiting time (difference between request and pickup time) in seconds
waiting time (90% quantile)	float	90% quantile for waiting time (difference between request and pickup time) in seconds
waiting time from ept	float	Average waiting time (difference between earliest pickup and pickup time) in seconds ($ \frac{1}{number_users} \sum_{i \in R_{served}} (\text{pick_up_time}_i - \text{earliest_pick_up_time}_i)$ )
detour time	float	Average detour time (difference between travel time and direct travel time) in seconds ($ \frac{1}{number_users} \sum_{i \in R_{served}} (\text{drop_off_time}_i - \text{pick_up_time}_i - \text{direct_travel_time}_i) - \text{boarding_time}$ )
rel detour	float	Average relative detour ($ \frac{100 [%]}{number_users} \sum_{i \in R_{served}} ( (\text{drop_off_time}_i - \text{pick_up_time}_i - \text{direct_travel_time}_i - \text{boarding_time} ) / \text{direct_travel_time}_i )$ )
% fleet utilization	float	Fraction of simulation time fleet vehicles are not idle
vehicle revenue hours [Fzg h]	float	Overall vehicle hours spent by vehicles actively serve bookings (states: boarding, waiting, route, repositioning)
rides per veh rev hours	float	The number of travellers divided by the revenue hours
rides per veh rev hours rq	float	The number of bookings divided by the revenue hours
total vkm	float	Total kilometres driven by all vehicles of the operator
occupancy	float	Distance weighted average number of passengers in the vehicle ($\sum_{i \in vehicle_driving_legs}N_i^{passengers} \cdot \text{driven_distance}i / \sum{i \in vehicle_driving_legs} \text{driven_distance}_i$)
occupancy rq	float	Distance weighted average number of bookings in the vehicle ($\sum_{i \in vehicle_driving_legs}N_i^{bookings} \cdot \text{driven_distance}i / \sum{i \in vehicle_driving_legs} \text{driven_distance}_i$)
% empty vkm	float	Percentage of km that vehicles drove around empty. This includes both repositioning (see next parameter) and empty trips when approaching to a booking.
% repositioning vkm	float	Percentage of vehicle km only used for repositioning
customer direct distance [km]	float	Sum of all served traveller distances on their corresponding fastest route
saved distance [%]	float	Meassure for pooling efficiency. This is computed in this way: $\frac{\text{customer direct distance} - \text{fleet vkm}}{\text{customer direct distance}}$
trip distance per fleet distance	float	This is computed in this way: $\frac{\text{customer direct distance}}{\text{total vehicle km}}$
trip distance per fleet distance (no reloc)	float	This is computed in this way: $\frac{\text{customer direct distance}}{\text{total vehicle km} * (1 - \% \text{ reposition vkm})}$
avg driving velocity [km/h]	float	Average vehicle velocity
avg trip velocity [km/h]	float	This is computed in this way: $\frac{\text{customer direct distance}}{\text{total travel times}}$ This can be useful for comparing travel speeds of different modes of transport
total toll	int	If there is a toll in a region, then this is the total toll that has been paid by the operator during the simulation
mod revenue	int	Sum of all booking fares
mod fix costs	int	Sum of fixed costs of the operator
mod var costs	int	Sum of variable costs of the operator. This is computed based on the driven distance
total CO2 emissions [t]	float	OUTDATED
total external emission costs	int	OUTDATED
parking cost	int	OUTDATED
toll	int	OUTDATED
customer in vehicle distance	float	Sum of kilometres travelled by costumers. Note that is is different to the distance that vehicles drove with a costumer in them: If two costumers shared a ride, this ride is counted double into this parameter.
shared rides [%]	float	Percentage of bookings that shared part of their ride with at least one other booking