11. Conclusions

11.2. First systems developed using Bridges technology

11.3. Looking ahead: a never-ending research

Bridges set out to fulfil the most demanding software requirements of an ideal ETIS but Bridges technology can also be applied for building up effective support systems in transport related fields (e.g. spatial development or environmental assessment) at various geographic levels (local, regional, national). This is demonstrated by the already available experience of implementing Bridges technology (see Section 11.2).
Bridges technology was designed as a set of highly interconnected tools: communication systems, data models and protocols, specialised transport routines, applications to build intelligent user interfaces, component software routines and stand alone applications etc. Each Bridges tool can work alone and none is indispensable, and optimum implementation will always be through integration with other tools and external applications. This approach is consistent with (and complementary to) current strategies adopted for management support systems and electronic business by leading software companies such as ORACLE and Software AG.

At the time of writing (late 1999), Bridges is a productive software technology which has been successfully applied to building of advanced support systems.

Systems developed using Bridges have the following advantages from the end-user point of view:

• They are scalable multi-software systems: supported by the Bridges Communication System, a Windows'NT/98 compatible technology, able to integrate multiple and specialised software applications, in particular sophisticated transport models, database managers and GIS.
• They are open to the integration of any Windows OLE/COM compatible application and component software. Advanced transport models (usually non-interactive applications, often not OLE/COM compatible) can be integrated through a new common data model and format (Bridges GTF), as well as transport database structures managed by GIS applications (Bridges GTF_GIS)
• They are highly productive systems, with a set of Bridges Core Utilities specialised in transport database management (e.g. network oriented database and GIS routines to harmonise information linked to heterogeneous graph structures).
• They are intelligent systems, able to facilitate intermediation between policy questions and outputs resulting from scientific models.

They are user friendly systems, with specific Bridges Core Utilities supplying the place of previously missing or sub-optimal data management and mapping tools needed to develop powerful user interfaces
From a system developer's point of view, Bridges technology has the following impacts:

• The time and cost of building up an advanced multi-software support system is dramatically reduced by Bridges Communication System (CS) (this explains why so many Bridges implementations have been possible even before completion of the research).
• Bridges Network Utilities (NIS) provide a number of independent modules, for instance new graph management tools (e.g. graph editing, matching, analysis etc.) which have already proved to be very efficient to develop precise and topologically rich graphs of Europe. This set of transport oriented database management tools is closely linked to transport modelling (including network assignment routines) and may be the starting point for building interactive models based on the formulation and results of more advanced external models.
• For the first time, Bridges data models and formats (GTF) provide a real possibility for transport modellers to exchange of databases. Currently, heterogeneous data models (even terminology) and individual formats make the data exchange process extremely costly, when it is feasible. This problem is blocking modelling research to some extent, since cross-fertilisation and interaction between models is very difficult (e.g. using the same database and comparison of results obtained with two different models is far from easy)
• Bridges data models and formats linking transport models and advanced (GTF_GIS), as well as the ArcInfo translator, allows use of GIS not only to display modelling results (as a mapping tool) but as a database management tool as well. These capabilities can become especially useful in a data warehouse approach.
• Bridges Directory of Transport Data Sources (DDG) demonstrates the actual feasibility of building simple and highly useful tools, with potential in the transport planning sector. DDG demonstrates the necessity of developing a set of electronic directories (for data sources, models, policy questions, experts and advisors etc.) in the ETIS context.
• Bridges Expert System (ES/DSS) is the first attempt to develop a methodology and, based on it, a software tool able to provide intelligent intermediation between advanced models and end users (decision makers). Complex tasks, such as writing model templates to interpret model results and translate user queries to model parameters, can be carried out relatively easily by system developers.

All considered, already existing experience proves that Bridges technology is an innovative software technology in the following senses:

• It brings new capabilities (e.g. new data models and formats to import data from advanced transport models, a new Communication System to drive many commercial applications without being dependent on any of them, new GIS utilities for network and transport database management etc.)
• It optimises and refines already existing software tools (e.g. mapping tools for transport networks, easier interfaces to key database management utilities, fully personalised interfaces etc.)
  

Many Bridges components are already being implemented by the Bridges partners, in collaboration with other companies and institutions. These experiences cover a variety of areas, from systems actually implemented in institutions and in use by their personnel (e.g. SIMU), systems in the process of being implemented (e.g. ICONgis -see www.mcrit.com/icongis-, BRAX, SIET etc.), acting as support to dissemination drives (Phare Toolbox) or educational initiatives (PIE), systems used in consultancy work by Bridges partners (PSCUT, SCANRAIL) or systems being developed within European research projects (e.g. Pyrenees, ATIS). Each one of these systems is different, uses different Bridges tools and commercial applications, and is at a different level of development. Despite the fact that not all Bridges tools have yet been used, especially the more advanced ones, all these support systems constitute excellent opportunities for testing and improving fundamental aspects of Bridges technology.

Even before the formal end of Bridges research, a number of real applications have been already implemented and work in hand of their users. For example the ICONgis/ EIB for the European Investment Bank to assess large transport infrastructure projects in Europe, and the SIMU/ IMU, Urban Information and Modelling System for the Institut Municipal d'Urbanisme of Barcelona, designed for environmental impact analysis at local level. These two policy support systems and several others under development helped to test software tools and validate the overall approach of Bridges. Even more important, they have been extremely useful helping Bridges technology to start the process of moving from the research laboratory to the of policy makers' and experts' personal computers.

In the context of technology oriented research such as Bridges it is always important to remember that a Policy Support System, or a more general Decision Support System (DSS), such as PJGIS or SIMU, is a software system under control of one or many decision makers that assists but does not replace them. Therefore, the development of a decision support system is about people, not about computers. Although computers and software play an integral role in the DSS world, the study of DSS is about how people think and make decisions. The definition and implementation of a DSS must integrate future users as much as possible, since for them a DSS represents both a challenge and an opportunity to improve their working processes. In any case, a DSS will induce organisational changes which cannot be successful unless they are clearly perceived and desired from the outset. Recent developments in the DSS field tend to integrate the multiple decisions being taken by the institution, so they become Organisational DSSs. A successful ODSS will be a participative rather than a mandatory process.

Because of these user and institutional requirements, only decentralised and highly interconnected modular technologies can support a DSS. Moreover, only modular technologies are flexible enough to be updated and improved according to rapidly and unexpectedly changing conditions for both information and technology.

One important implementation advantage of Bridges technology is that being 100% financed by EC it is royalties free for all EU institutions (and others with EC agreement), and therefore an unlimited number of users can be connected to systems developed on Bridges with no licensing fees. For large institutions this advantage is crucial at least in the short term to stimulate their interest to be involved in the development of their own support systems, a process which anyway would require a serious organisational effort. The following list exemplifies the on-going experience in building up friendly and open multi-software systems based on Bridges technology (in some cases including decision support capabilities, in other cases just modelling and/or information):

1. ICONgis in the European Investment Bank (Transport Infrastructure Assessment, May 1999-October 1999)
2. SIMU in the Urban Planning Department of the City of Barcelona (Environmental Impact, January 1998-October 1999)
3. Phare Toolbox for DG7/Phare Countries (Dissemination of data, formulation and results of traffic forecast models, July 1999- end planned November 1999)
4. BRAX in the Barcelona Regional Agency (Urban Development linked to Public Transport, starting October 1999- end first phase February 2000)
5. PSCUT for JAE Junta Autostradas de Portugal (Mapping/Strategic Modelling, January 1999-July 1999)
6. SIET for the Metropolitan Plan of Barcelona (GIS/Strategic Transport Modelling, February 1998- January 2000)
7. SCANRAIL-Denmark (GIS/Transport Modelling links through Bridges ArcInfo translator, July 1999)
8. ATIS (Alpine Policy Information System demonstrator) for DG7 as a pilot case-study for ETIS (December 1999)
9. Pyrenees Information System (for ASSEMBLING research, 4th EU FP, October 1999- July 2000)
10. PIE GIS/Mapping system for educational purposes (Educational Software Program, Catalonia). Free GIS/Mapping for 2,000 schools (500,000 students between 16-18 years).

These applications (for further information please visit www.mcrit.com) demonstrate the feasibility of using Bridges technology to build up advanced policy support systems in transport or other closely related policy fields, as well as their productivity.

The Bridges research effort, by its very nature, could not reach a final, definitive conclusion. Each one of the systems mentioned requires the improvement of existing tools and even, in same cases, the development of new ones. For example, the "bridges" between transport models and Virtual Reality and 3-Dimensional applications, which were not part of the original Bridges research because their limited interest for European transport planning, are more important when developing support systems at urban scale. Other "bridges" with advanced modelling tools such as GAMS, TRIO, SPSS etc. will also be of greater interest and are expected to be developed as required by new applications). More importantly, many advanced Bridges technologies (e.g. GTF format for data exchange, and Expert System/DSS) have not yet been applied in real cases. It is expected that many of these more advanced technologies will be taken forward in further research activities.

Finally, the development of a "bridge" between Intranet multi-software systems build with Bridges and Internet is the area of most interest from a software development point of view. In the context of ASSEMBLING, another 4th Framework Programme project, which aims to develop a pilot Internet executive support system for European transport policy-makers, this is currently being explored. Once achieved, the integration of the results of both projects may provide for an optimum use of forefront software innovation to support a friendly access (both Internet and Intranet) to advanced transport information and modelling systems.

info@mcrit.com