The basic idea behind the service is to virtually connect existing travel information systems. The former isolated individual systems are linked via central components so that continuous information can be calculated. Optimisation technologies are used in order to fulfil customer requirements. The customer can enter the whole international system via the Internet by contacting any of the participating information systems on which all information can be displayed.
The independent local systems are connected via central components which are needed for the generation of itineraries. These central components are:

• RODI – Ring origin destination identificator. The RODI tries to match the user input to start and destination locations. In order to do so, the RODI contacts the appropriate passive servers.
• RCC – Ring connection composer. This RCC acts as a super connection composer and retrieves and combines the partial information by open interfaces to the formerly isolated information systems that now act as passive servers.
• RRDB – Ring Reference database. At this database – among others – all transition stations are stored. Transition stations are points where an interchange between two local systems or between a local system and an interregional system (e.g. national railway) is possible.

The local systems keep their user interface (GUI), algorithms and database structures and are capable to display international itineraries in their local format.

Background process
The background process is the maintenance and update of all central data that are required as meta data in order to generate itineraries. The process covers the definition or redefinition of central data. This data is stored in the RRDB and consists of:

• list of city and town names within the participating regions
• Information about participating servers
• Harmonised data that are necessary to meet the customer demand (e.g. selection of train categories and symbol codes)
• Transition points (nodes where different partial itineraries from the participating information systems must be connected in order to retrieve optimal itineraries)

On-line processing
On-line processing is the generation of itineraries according to a customer request. The following steps are taken by the system after the user has put a request at any of the participating local systems:

1. The start and destination servers are determined by the RODI. Country, region or city names will be used for identification at the central database.
2. The input of the customer is verified by the identified passive servers. If the requested stop/address matches more than one stop/address, the customer needs to select the right stop/address.
3. The departure and destination information is transferred to the RCC.
4. The RCC contacts the needed passive servers (information systems) requesting all necessary transition points.
5. Then, the RCC sends requests to all involved passive servers to generate partial itineraries (from departure or destination point to all local transition points). This is done in the following order: first, the long-distance part is generated – based on this information the local servers are contacted.
6. The RCC composes these partial itineraries and generates total itineraries which best fit the customer’s criteria.

Technical benefits of the EU-Spirit service
In general, every operator – regardless of the algorithms used – can participate in this open service system. Any updates of the time schedules in the local systems are automatically available for the EU-Spirit system which reduces the maintenance costs and improves the accuracy of the system and the information it provides.