User Manual

Welcome to the documentation section. You should find all the information you need about the Travel Simulator in these pages. To setup and get started with theTravel Simulator or one of the modules have a look at the getting started page. For step by step instructions and exemples on how to use all the modules the tutorials page is what you are looking for.

For a quick and high level description of Travel Simulator and all the modules, you are in the right place. A list of the Travel Simulator's modules and a brief description of them is available in the repositories list.

The dependencies diagram does the tought job of showing you whom needs what.

Last but not least come the simulation input files that are mandatory for some of the Travel Simulator modules.

Still have a question? The F.A.Q. is here.

Repositories List

TvlSim: C++ Airline Market Simulator. TvlSim uses TraDemGen to generate booking requests, then calls SimCRS to build the corresponding travel solution list (a travel solution is made of a segment path obtained from the schedules, as well as the corresponding prices and availabilities), asks TravelCCM to make a choice among the travel solution list and finally comes back to SimCRS to make a potential sell.

TraDemGen: C++ Travel Demand Generator. Takes a demand input file and generates a stream of requests which uses Event Queue as a temporary storage. In the input file, for each origin-destination and each day, the mean of demand requests number, the trip type (round-trip or one way), the probability to accept restrictions, the willingness to pay, the preferred arrival time, etc. are defined.

SimCRS: C++ Simulated Computer Reservation System (CRS). Plays an intermediary role in the process: calls AirSched to build travel solution lists, calls SimFQT to compute prices, calls AirInv to get availabilities or to make a sell, ...

AirSched: C++ Simulated Airline Schedule Service. AirSched takes care of creating schedules from an input file which contains information such as airline code, flight number, departure date, legs (a leg is a direct flight and a flight can contain several legs), boarding time, etc.

AirInv: C++ Simulated Airline IT Inventory System. AirInv uses schedule input file to manage airlines Inventories: for example it is called by SimCRS to compute availabilities.

RMOL: C++ Revenue Management (RM) Open Library. Contains Revenue Management algorithms.

AirRac: C++ Airline Revenue Accounting (RAC) library. Generates a yield cache thanks to a yield input file.

SEvMgr: C++ Simulation-Oriented Discrete Event Management Library. SEvMgr handles the Event Queue containing all the simulation events. TraDemGen pushes booking requests and cancellations into the queue whereas AirInv adds Inventory snap shots and revenue management events. This Event Queue is the representation of the time, each Event (snapshot, booking request, cancellation...) is sorted according to the time.

SimFQT: C++ Simulator Fare Quote (FQT) Library. Contains all the pricing algorithms to help fill in the travel solutions with the appropriate price.

TravelCCM: C++ Travel Customer Choice Model (CCM) library. TravelCCM plays the role of a customer by choosing the best travel solution from a provided list. The choice can be based on the price, the time constraints, the restrictions...or a mix of all those parameters. This part can be modified in order to simulate at best passenger behaviours.

StdAir: C++ Standard Airline IT library. This library is used by all others modules, it contains all business objects useful in Airline IT (for example: Inventory, Fare Family, Booking Class, Booking Request, Cancellation, ...).

AvlCal: C++ Simulated Availability Calculator Library.

SimLFS: C++ Simulated Low Fare Search (LFS) System.

MetaSim: Helper containing all simulator components.

SimRDS: Simulation Reference Data Sets.

Global-Documentation: General documentation for the overall project and community.

Demand-Modelling: Demand Modelling studies.

Dynamic-Pricing-Studies: Dynamic Pricing studies.

Travel Simulator: this very website.

Dependencies diagram

Simulation input files

Here is a short description of each input file you may have to create when dealing with one or more Travel Simulator modules.

The demand input file is composed of one or more lines, each line representing a demand stream:

2009-01-01; 2011-03-03; 1000000; SIN; BKK; Y; SIN:0.7, BKK:0.2, row:0.1; DF:0.1, DN:0.3, IF:0.4, IN:0.2; RO:0.6, RI:0.2, OW:0.2;
0:0.1, 1:0.1, 2:0.15, 3:0.15, 4:0.15, 5:0.35; P:0.01, G:0.05, S:0.15, M:0.3, N:0.49; 0.5; 50; 0.5; 50;
06:0, 07:0.1, 09:0.3, 17:0.4, 19:0.8, 20:0.95, 22:1; 160; 15:0, 60:1; 330:0, 40:0.2, 20:0.55, 10:0.75, 1:1; N, 49.0, 7.0;

Let's have a look at each attribute of the demand stream above.

Start date of the departure date period (yyyy-mm-dd)
End date of the departure date period (yyyy-mm-dd)
Day of the week

2009-01-01;
2011-03-03;
1000000;

Those three attributes are used to compute the list of dates at which passengers are willing to travel. In this example, we are going to generate booking requests for each Monday between 2009-01-01 and 2011-03-03.
For this demand stream, passengers wish to travel from a certain origin to a certain destination within a certain cabin:

Origin (3-char airport code)
Destination (3-char airport code) 
Preferred cabin ('F' for First, 'C' for Business, 'Y' for Economy)

SIN;
BKK;
Y;

The next parameters define distributions with Probability Masses:

Point-of-sale (row stands for rest of the world)
Channel (D=direct, I=indirect, N=online, F=offline)
Trip type (RO=outbound of round-trip, RI=inbound, OW=one way)
Stay duration (number of days)
Frequent flyer (P=Platinum, G=Gold, S=Silver, M=Member, N=None)
Change fees restriction and its disutility value
Non refundable restriction and its disutility value

SIN:0.7, BKK:0.2, row:0.1;
DF:0.1, DN:0.3, IF:0.4, IN:0.2; 
RO:0.6, RI:0.2, OW:0.2; 
0:0.1, 1:0.1, 2:0.15, 3:0.15, 4:0.15, 5:0.35; 
P:0.01, G:0.05, S:0.15, M:0.3, N:0.49; 
0.5; 50; 
0.5; 50;

Then we encounter parameters defined as continuous cumulative distributions:

Preferred departure time (hh.mm)
WTP (moneraty value)
Value of time
Arrival pattern (DTD as a positive value)

06:0, 07:0.1, 09:0.3, 17:0.4, 19:0.8, 20:0.95, 22:1;
160
15:0, 60:1; 
330:0, 40:0.2, 20:0.55, 10:0.75, 1:1;

Finally, we find the type of distribution followed by the number of requests to be generated for each specified departure date:

Distribution of number of requests

N, 49.0, 7.0

The schedule input file is composed of one or more lines, each line representing a flight period:

SQ; 12; 2009-01-01; 2011-03-03; 1001000; SIN; BKK; 08:30; 11:10; 03:40; Y; 200; 0; Y; YBM; 1; F1; D1; YBM;
SQ; 12; 2012-04-27; 2012-05-03; 1000000; SIN; BKK; 09:20; 12:00; 03:40; Y; 150; 0; Y; YBM; 1; F1; D1; YBM;

The first parameters of each line describe a specific schedule:

Airline code
Flight number
Start date of the departure date period (yyyy-mm-dd)
End date of the departure date period (yyyy-mm-dd)
Day of week
Origin (3-char airport code)
Destination (3-char airport code) 
Boarding time (hh.mm)
Landing time (hh.mm)
Elapsed time (hh.mm)

SQ;
12;
2009-01-01;
2011-03-03;
1001000;
SIN;
BKK;
08:30;
11:10;
03:40;

In this example, Singapore Airline (airline code "SQ") is operating a flight numbered 12 from Singapore to Bangkok each Monday and Thursday between 2009-01-01 and 2011-03-03.
The remaining parameters deal with cabin, capacity, fare family and booking classes:

Cabin ('F' for First, 'C' for Business, 'Y' for Economy)
Cabin capacity
Segment specificity ID
Cabin code
Class code List
Fare Family ID
FRAT5 Curve ID
Fare family Disutility Curve ID
Class code List

Y; 
200; 
0; 
Y; 
YBM; 
1; 
F1; 
D1; 
YBM;

The fare input file structure should look like the following sample:

1; SIN; BKK; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; SIN; Y; IN; 7; T; T; T; 3; 150.0; SQ; Y;
2; SIN; BKK; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; BKK; Y; IN; 7; T; T; T; 3; 150.0; SQ; Y;
3; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; SIN; Y; IN; 7; T; T; T; 3; 150.0; SQ; Y;
4; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; HND; Y; IN; 7; T; T; T; 3; 150.0; SQ; Y;
5; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; ROW; Y; IN; 7; T; T; T; 3; 150.0; SQ; Y;
6; SIN; BKK; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; SIN; Y; IF; 7; T; T; T; 3; 150.0; SQ; Y;
7; SIN; BKK; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; BKK; Y; IF; 7; T; T; T; 3; 150.0; SQ; Y;
8; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; SIN; Y; IF; 7; T; T; T; 3; 150.0; SQ; Y;
9; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; HND; Y; IF; 7; T; T; T; 3; 150.0; SQ; Y;
10; SIN; HND; OW; 2010-01-15; 2010-12-31; 00:00; 23:59; ROW; Y; IF; 7; T; T; T; 3; 150.0; SQ; Y;

Each line represents a fare rule, identified by the first number of the line:

Number of the fare rule
Origin (3-char airport code)
Destination (3-char airport code)
Trip type ('RT'=round-trip and 'OW'=one way))
Start and End date of the validity period (yyyy-mm-dd)
Start and End time of the validity period (hh:mm)
Point-of-sale
Chosen cabin ('F' for First, 'C' for Business, 'Y' for Economy)
Channel ('D'=direct, 'I'=indirect, 'N'=online, 'F'=offline)

1;
SIN;
BKK;
OW;
2010-01-15; 2010-12-31;
00:00; 23:59;
SIN;
Y;
IN;

The next parameters define the fare and its corresponding fare options:

Advanced purchase (number of days)
Saturday night stay option ('T' stands for true and 'F' stands for false)
Change fees option ('T' stands for true and 'F' stands for false)
Refundable criterion ('T' stands for true and 'F' stands for false)
Minimum stay duration (number of days)
Fare amount

7;
T;
T;
T;
3;
150;

Finally comes the airline and the class for which the fare is applicable:

Operating airline code
Chosen booking class

SQ;
Y;

To sum up, each line of the fare input file tells us the price a customer will be asked to pay according to a lot of criteria such as:

the origin and destination of his travel (for instance from Singapore to Bangkok in the first fare rule).
the type of his travel.
the date and time he is willing to travel (each fare rule has a date range and a time range of validity).
the place where he is buying the ticket, i.e. the point of sale ("ALL" meaning all points of sale).
his chosen cabin.
the channel of the booking described by a two letters code ("DC" meaning all different channels).
the date when he wants to buy the ticket, i.e. the advanced purchase required in number of days.
the saturday night stay option, i.e. is he staying a Saturday night between his inbound trip and his outbound one?
the change fees option, i.e. are there fees to change his ticket?
the refundable criterion, i.e. is the ticket refundable?
the number of days he is willing to stay at the destination location (each fare rule has a minimum stay requirement in number of days).
the operating airline and the chosen booking class.

The yield input file structure should look like the following sample:

1; SIN; BKK; OW; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 400.0; SQ; Y;
2; SIN; BKK; OW; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 250.0; SQ; B;
3; SIN; BKK; OW; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 160.0; SQ; M;
4; SIN; BKK; RT; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 500.0; SQ; Y;
5; SIN; BKK; RT; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 300.0; SQ; B;
6; SIN; BKK; RT; 2009-01-01; 2012-06-01; 00:00; 23:59; ALL; Y; DC; 200.0; SQ; M;

Each line represents a yield rule, identified by the first number of the line:

Number of the yield rule
Origin (3-char airport code)
Destination (3-char airport code)
Trip type ('RT'=round-trip and 'OW'=one way)
Start and End date of the validity period (yyyy-mm-dd)
Start and End time of the validity period (hh:mm)
Point-of-sale ('ALL' meaning all points of sale).
Chosen cabin ('F' for First, 'C' for Business, 'Y' for Economy)
Channel ('DC' meaning all different channels)

1;
SIN;
BKK;
OW;
2009-01-01; 2012-06-01;
00:00; 23:59;
SIN;
Y;
IN;

The next parameter is the yield value:

Yield amount

400;

Finally comes the airline and the class for which the yield is applicable:

Operating airline code
Chosen booking class

SQ;
Y;

To sum up, each line of the yield input file tells us the value (useful for the revenue management system) a customer will be given according to a lot of criteria such as:

the origin and destination of his travel (for instance from Singapore to Bangkok in the first yield rule).
the type of his travel.
the date and time he is traveling (each yield rule has a date range and a time range of validity).
the place where he is buying the ticket, i.e. the point of sale.
his chosen cabin.
the channel of the booking described by a two letters code.
the operating airline and the chosen booking class.