ABC'2000: Airline Business ComputingAdvanced Development ProjectsRaCoSy: Railway Computing SystemsReturn to UNU/IIST's home page

II/1/2/1 RaCoSy: Railway Computing Systems

Synopsis

This research and development project is concerned with an overall determination of a normative software architecture that will allow co-existence of, data exchange between and mutual invocation among arbitrary Railway Computing Systems (hence RaCoSy) software packages.

Period

April 1993 -- March 1996

Partners

Chinese Ministry of Railways

Aims & Objectives

To enable own development of arbitrarily advanced software systems for own consumption, human resource development, and potential international commercialisation of railway software. In the past as now, most software for developing countries' railway needs is acquired from industrialised countries. The RaCoSy project proves that this need not necessarily continue to be so in future. In particular:

  1. Joint research of infrastructures and normative software architecture components for the Chinese railway system

  2. Training of Fellows from the Chinese Ministry of Railways in advanced software technology development

  3. Joint development of selected software packages: Distributed Train Dispatch (initially for the Zhengzhou-Wuhan line), and Station Management.

Achievements

During 1994 four Fellows joint with UNU/IIST staff and visiting experts developed a set of domain analysis, requirements capture, software architecture and design descriptions for a prototype single standing (i.e. non-distributed) Running Map train rescheduling system. This Sun/UNIX-based tool has now been delivered to the Zhengzhou Railway Administration in China by returning Fellows and a three week training course on its design and use was given mainly by these Fellows in December 1994.

During mid-94 to mid-95 one Fellow did the domain analysis, requirements capture software architecture and partial implementation of a Station Management system.

Two new Fellows from China were intended to come in January 1995 but were not able to arrive until August 1995. Two problems were tackled with these Fellows: developing a distributed running map tool (since actual dispatching is handled by distributed dispatch units) and improving on the simplified model of dispatching used by the prototype tool. They had to return to China in March 1996 before this work was completed.

During the academic year 1995-6, UNU/IIST collaborated with the Chinese University of Hong Kong in a project to take the running map tool further -- into a tool that would apply constraint propagation techniques to the rescheduling problem, and suggest solutions to the dispatcher rather than merely validate proposals. Jimmy Ho Man Lee and Ho-Fung Leung together with two final year undergraduates and a PhD student produced, from the UNU/IIST RAISE specifications, a tool that reproduced the running map prototype and also provided optimality criteria and heuristics to tackle rescheduling.

Reports

A number of UNU/IIST Research Reports (III/1) relate to this project: 23, 37, 42, 52, and 93.

In addition a large number of project reports are available


iistinfo@iist.unu.edu, December 1996

ABC'2000: Airline Business ComputingAdvanced Development ProjectsRaCoSy: Railway Computing SystemsReturn to UNU/IIST's home page