Implementation of Real-Time Data into a Transport Simulation

Transcription

1 Implementation of Real-Time Data into a Transport Simulation Krain, Armin Johannes Deutschmann, Andreas Rudolph, Florian German Aerospace Center Abstract Within the framework project Optimode.net the German Aerospace Center works on a more efficient cross-linking of different modes of transportation while focusing on aviation. This project contains a traffic simulation and a control station environment to improve the travel chain of every passenger. In earlier versions, the system was based on a database containing fictitious data. Therefore, a tool was developed, which extracts data from Web-APIs and integrates them into the project s database. The tool is designed to be upgradeable with additional APIs and customizable to the structure of the database. Introduction In 2011 the Advisory Council for Aviation Research and Innovation in Europe (ACARE) published aims for aviation in Europe by the year One of these aims states that 90% of passengers within Europe should be able to reach their destination in a time frame of four hours (1). This not only considers the flight, but the whole door-to-door journey. Therefore, new concepts of transport management are required to improve the cross-linking of different modes of transport and shorten transfer times. As an important junction for transport for a journey including a flight, we focused on the airport as a part of the passengers travel chain. The airport plays an important 1

2 role for the improvement of this travel chain out of two reasons. On the one hand, several modes of transport interfere at airports. Besides the air traffic, most airports offer connectivity via road and rail. In 2015 over 35 million passengers departed or arrived at Germany s biggest airport in Frankfurt via ground-based transport modes such as private cars, taxis, rental cars or long- and short-distance-trains (2). Frankfurt airport can be accessed via a long-distance-train station with over 170 connections daily (3). On the other hand, passengers spend much time on their way from the airport entrance to the plane, while being processed in the following steps (4). If the passenger has not checked in for his flight online, he has to visit a check-in counter or a selfservice kiosk. In case he travels with checked baggage his next station is a baggage drop-off point. The latest time of baggage drop-off and check-in is determined by the airline and often depends on the departure airport, destination and booked class. For instance at Chicago s O Hare airport, Lufthansa and United counters close 60 minutes before departure time for international travelers with checked baggage. For domestic flights with carry-on luggage only, counters remain open 30 minutes prior departure. In Munich Lufthansa s business and first class passengers need to check-in 30 minutes prior departure, regardless if they travel with checked baggage or internationally (5,6). Travelers may experience longer lines at Transportation Security Administration (TSA) checkpoints at U.S. airports. The TSA advises to arrive two hours before the flight for domestic travel and three hours before for international travel plus extra time during peak hours (7). After passing the security checkpoints, passengers need to get to their gate and wait for boarding. The aim of Optimode.net is not only to shorten this transfer time, but also to decrease the incidence of passengers missing their flight. This is done by determination of the transfer times between the checkpoints. 2

3 Framework Project Optimode.net The German Aerospace Center develops an interactive control station environment for traffic purposes, named Optimode.net. Optimode.net creates a trajectory for each passenger based on possible transport connections and preferences of the traveler. This trajectory contains the current travel chain and can be changed before or while the passenger is traveling. Therefore, the whole journey from door to door and as many transport modes as possible have to be considered. In the current stage of Optimode.net every travel chain ends with a flight departure from the Generic International Airport (GIA), a fictitious model of an international airport with approximately 13.5 million passengers and 160,000 flight movements per year (8). All data from passengers and transport connections are collected in a central data hub and used for the calculation of the trajectories. In case of delays or cancellations of connections it is possible to reroute the passengers, even to connections of other transportation modes. The aim is that every passenger reaches its destination on time. For the time being, Optimode.net uses a passenger simulation software to model passenger behavior. For the transport connections fictitious data are used. To improve the quality of the simulation, a tool FRAtoOptimode was developed, which extracts information from online data interfaces and adds them to Optimode.net. Such interfaces are called APIs (Application Programmable Interfaces). They are used to transfer data on a machine-readable basis. Along the trend of Open Data, the number of available APIs is rising (9). Interfaces for aviation can be divided into three groups by their publishers: Transportation facilities, transportation service providers and third-party services. With Frankfurt Airport and Lufthansa, the biggest German airport and airline offer APIs for flight data, which are free of charge. For railway data, Deutsche Bahn announced to extend their Open Data system as only scheduled departure and arrival times of long distance trains are published currently. 3

4 Methods The Frankfurt Airport publishes its flight data via a Web-API. The interface delivers data from flights which land or take-off in Frankfurt, formatted in the Javascript Object Notation (JSON). It is designed as a REST-ful web service and offers several functions which can be accessed via GET requests. FRAtoOptimode requests the departure and arrival flight plan of the current day, converts the flight data and inserts them into the MySQL database of Optimode.net. The data type structure is depicted in Figure 1. For the development of this console application, Visual C# was used. Program settings and information to connect to the database and APIs can be made in a XMLfile, which is located in the program folder. The file is structured self-descriptive and its contents can be changed with most common text editors. It is designed to be easily expandable with new APIs. The tool itself consists of three parts: Main, ApiHandler and Translator. In the Main part the settings file is loaded and handlers for the APIs are initialized. The class API Handler is completely independent of the APIs structure. It provides functions for reception of API data, connection and update to the Optimode.net database and conversion of API data into C# internal objects. The Translator contains API specific functions. It is supplied by the structure files of the interfaces and translates the data API-specific into the structure of the database. Some APIs require additional functions Figure 1: Data types of FRAtoOptimode 4

5 for example for requesting additional data or getting access to more datasets. These functions are also implemented in the Translator. Figure 2 shows the program structure. This structure allows an implementation of additional APIs to the tool without changes in the API Handler. The API-specific structure files can be created automatically, changes are only necessary in the Translator, Main class and the settings file. Figure 2: Program structure of FRAtoOptimode 5

6 Conclusion The tool FRAtoOptimode was successfully tested in a simulation environment at the German Aerospace Center. It uses APIs to enlarge the data sources with real-time data from the Frankfurt Airport and scheduled arrival and departure times for long distance trains of the German railway network. More interfaces are planned to be added to the software in the future. To gain independence from the APIs accessibility, example datasets, extracted from the interface, are saved locally and can be used as soon as the API is inaccessible or reuse them for different simulation scenarios e.g. a bad weather day. Problems resulted from the quality of the APIs. There have to be available APIs which include the data needed for the simulation. Occasionally, incorrect or doubled data were delivered by the interfaces. For doubled data a filter mechanism was implemented which only implements the first data entry into the simulation database. Structural changes of the interfaces can cause the necessity of a change of the program code of FRAtoOptimode. Web-APIs offer a way to implement data into a transport simulation. Without the integration of real-time data, an application of the simulation environment Optimode.net to the reality is not possible. The integration of real-time flight and scheduled train departures and arrivals via FRAtoOptimode is a first step to gain applicability. With the combination of multiple APIs, the representation of more complex traffic models will be achieved. 6

7 References 1 : ACARE; 2011; Flightpath 2050, Europe s Vision for Aviation. 2 : FRAPORT AG; 2016; Frankfurt Airport Air Traffic Statistics 2015; http: // raffic-figures/traffic-figures-more-information/air-traffic-statistics- 2015/jcr:content.file/air-traffic-statistics-2015.pdf 3 : FRAPORT AG; 2015; Zahlen, Daten, Fakten 2015 zum Flughafen Frankfurt; ht tp:// /2015/zahlen--daten--fakten-2015/jcr:content.file/zahlen-daten-fakten pdf 4 : Mensen, H.; 2013; Planung, Anlage und Betrieb von Flugplätzen; Springer; Berlin Heidelberg. 5 : Lufthansa Homepage; Check-in Annahmeschlusszeiten und Gepäckabgabefristen; accesed ; tionservice/checkin/checkininformation/!ut/p/c5/pzdlcomweew_pv-qvg1cl0k ax6aiyay6eqtfld66kc39-yz07vdozpjwl5ddwmj_6z_j0d_gdeknupowdaryh6syfooqsyqk KcNcVVIb3_KGdYKa4hBFqgSIAyr8Ks1MqKU-eS69QSHi3zTdOEF_SW9sy3jwx3LXjqdT0g7Te rggzs4z0srqrlokc-fqccwsutvderzynfupebzov9kq-wwa9dsky3ljl7h7aicaqrm!/dl3/d 3/L3dPQSEvUUt3QS9vQmch/?l=de&cid=18002&p=LH&s=DE 6 : United Homepage; Check-in and airport processing times; accessed ; 7 : TSA homepage; Frequently Asked Questions; accessed ; ww.tsa.gov/travel/frequently-asked-questions 8 : Milbredt, O., Rudolph, F. and Grunewald, E.; 2016; Passenger-centric Intermodal Traffic Management involving Airports and Railways; Canadian Transportation Research Foundation; Toronto. 9 : Mork-Ulnes, N.; 2016; Beyond;