DECISION SUPPORT SYSTEM FOR THE DEVELOPMENT OF RURAL WATER SUPPLY SCHEMES

DECISION SUPPORT SYSTEM FOR THE DEVELOPMENT OF RURAL WATER SUPPLY SCHEMES
Report No 837/1/01

EXECUTIVE SUMMARY

The report represents the final documentation of WRC project K5/837, 'Guidelines for the Development of Rural Water Supply Schemes - Further Development of a Decision Support System' which was a continuation of a previous project that concluded with WRC Report 677/1/98 by Sami and Murray. The original project ('Guidelines for the Evaluation and Development of Water Resources for Rural Development with an Emphasis on Groundwater') was designed to integrate many of the perceived issues related to the development of rural water supply schemes and to develop some new approaches to assessing groundwater resources and their potential for exploitation within relatively small scale rural water supply schemes. The project reported on here was designed to establish a computer based Decision Support System (DSS) and to populate it with the information and approaches outlined in the earlier report. The primary objectives of the project were :

To setup and develop the Delphi shell DSS for application to any subject requiring interactive decision-making computer support.
To create and develop the Rural Water Supply DSS for use within a Windows application.
To promote the use of the RWS-DSS for planning and evaluating future rural water supply projects.

Supporting objectives were:

To fill in more detail related to some issues than was possible during the first project.
To test the applicability of the DSS using established and new rural water supply schemes.
To test the usefulness of the DSS as a training tool for enhancing the awareness of various practitioners in he field of the many issues influencing the success or failure of rural, water supply schemes.
To hold workshops to present the DSS to established practitioners and obtain feedback on its advantages and disadvantages.

The project was designed to begin in January 1997 and to be led by the authors of the first report. However, the start was delayed due to a change in project personnel, a situation that also contributed to a slower start than was originally envisaged as this meant that the new project leader had to spend time becoming familiar with the approach adopted in the first project, as well as the contents of the report. During the course of the project, the approaches used by the Department of Water Affairs and Forestry (and others) in the design and implementation of rural water supply schemes were also being radically revised, in keeping with the social and political changes that were occurring in post-Apartheid South Africa. This meant that many of the institutional issues related to rural water supply only started to be properly defined during the project. The difficulties of developing the content of the DSS under such dynamically changing conditions needs to be stressed. A further point is that, on reflection it was not a very good idea to develop the generic DSS computer code at the same time as establishing the content of the specific Rural Water Supply DSS. Corrections to the code and associated procedures for entering the RWS information meant that old ground had to be re-covered quite frequently, slowing down the rate at which further information could be added, or existing information content tested. Despite all these problems, the project has succeeded in developing a satisfactory DSS shell, transferring the information contained within the original report (WRC Report No. 677/1/98) to the DSS, adding more up-to-date information and approaches, as well as addressing some of the objectives related to testing and training. There do, however, remain a number of issues associated with further technology transfer and the assessment of the system for solving practical problems that the project has been unable to address. The project team, in cooperation with the WRC and the Project Steering Committee are making every effort to deal with these problems as part of the post-project technology transfer actions.

The report is divided into five sections so that readers with different levels of interest can easily find the relevant parts. The first section (A: Project Context) reports on the background to the project, lists the original objectives and potential benefits, as well as providing a non-technical summary to the project as a whole. It also contains the details of any tests of the developed system that have been carried out, technology transfer actions that have taken place, as well as the main conclusions and recommendations for future work.

The second section (B: User Guide: Rural Water Supply Decision Support System - RWS-DSS) represents a guide to the use of the rural water supply DSS that has been developed. This section provides an overview of the DSS and a guide to its use, rather than being a complete reproduction of the content of the DSS itself, much of which was already covered by the earlier report (WRC Report No. 677/1/98). The objective of this section is to provide a hard copy framework that should assist the potential user in the application of RWS-DSS and which highlights some of the issues that users should be aware of. Section B is designed in the format of a supplementary manual to accompany the guidelines contained in WRC Report No. 677/1/98. Wherever that report has been updated by additional material more detail is provided within this text. Sections B4 to B 11 are written to guide the RWS-DS S software user through the various components and sub-components contained within the computer program and to lead the user to the more detailed step-by-step guidelines built directly into the RWS-DSS computer program itself. The respective sections are presented in the recommended order suggested for application in a decision support environment for rural water supply and the user is encouraged to adopt the approach presented in this step-by-step sequence.

The third section (C: DSS Shell System) explains the approach used in the design of the generic DSS Shell (written in Delphi code) and is designed as a manual for those users who might be in a position to further develop the RWS-DSS system, or who might wish to apply the shell to a different problem. It provides the full technical details about how to establish the text files that populate the DSS Shell with information, numerical inputs, calculations and flow control, etc. for a specific DSS.

The fourth section (D: RWS-DSS Structure. and Contents) lists (in a structured way) all the components of RWS-DS S and includes cross references to the text input files that deal with those components. This section can be used to obtain an overview of the content of the system, as well as directing a future developer (i.e. someone who may wish to update the system) to the relevant text input files. The content of this section can be used with a utility program (DSSViewer) to access the detail of the system design interactively.

The fifth section (E: Glossary of Terms) provides a glossary of the terms used within the report and has been included to ensure that there is as little confusion or misunderstanding as possible, given that different disciplines (and even different individuals within the same discipline) often use different terminology for the same thing.

The RWS-DSS computer program has been based on rural water supply project planning and implementation in the context of the South African Reconstruction and Development Programme. The computer program is generic in content but designed to encourage the user to consider all factors affecting rural water supply provision, bring together information and implement technically and economically appropriate rural water supply systems which are socially acceptable. The initial target user group can be broadly defined as personnel or institutions working on projects implemented by the Department of Water Affairs and Forestry (DWAF) within South Africa. However, it is likely that the system could also be of value to rural water supply developers in other parts of Southern Africa where the issues and problems are broadly similar.

The main recommendation is that the system should now be put to use, thoroughly tested and modified where necessary to ensure that it realises its potential for the solution of real problems. This means that it has to move away from the research environment of a University and into the practical environment of a government department or private consultant. One of the post-project tasks of the project team, with the cooperation of the WRC, is to find suitable custodian organisations for the system who have the resources to fulfill this role.