Environment and Planning B: Urban Analytics and City Science

The coordination of basic infrastructure, housing, and community facilities is important in new town development in order to ensure that all essential facilities are available to new residents on arrival. A microcomputer-based land information system (LIS) has been used to examine the applicability and usefulness of LIS in the estimation of land availability and population which are essential for the programming and monitoring of new town development. The capability of the microcomputer in managing the LIS is also examined.

The most notable problems of rapid urbanisation in Malaysia today include the need for land, housing, and provision of services. The link with day-to-day planning problems, however, remains a critical factor. The quality of planning and decisionmaking processes can be substantially improved when valid data are appropriately and efficiently handled. The aim of this paper is to demonstrate how the application of geographic information systems (GIS) can aid urban land-use planning and monitoring at the local authority level in Kuala Lumpur, with particular reference to a case study of low-income housing development. Data requirements, problems of data availabilitiy, and the appropriateness of different analytical tools are examined, followed by discussion of the feedback between the information system and the planning-decisionmaking process which is necessary for the successful implementation of new GIS technology.

Remote sensing (RS) and geographic information systems (GIS) can provide essential support for each other's operations. RS can provide information on current and changed land-use cover for a GIS, and, more importantly, a GIS can provide an essential expert knowledge base to help automate the technically demanding aspects of RS change-detection analysis. No commercially available system provides effective support for this important synergistic relationship. A prototype integration of both RS and GIS processing in association with an expert system has been developed. Vector and raster data are stored in a unified database which is accessed by an expert system to control key aspects of RS change detection. New RS interpretations are accumulated in the common database, providing potential for developing a system with a ‘learning’ function. The results show there is good promise for reducing some of the technical aspects of RS change analysis which have limited its application by practising planners, and for providing a significantly improved regional analysis tool.

It is clear that geographic information systems (GIS) offer many new opportunities for both geographic research and planning practice. The problem is how to cope with these possibilities and how to resolve the various technical problems which beset the present generation of GIS. Two fundamental problems of GIS will be discussed. On the one hand, GIS systems are not advanced enough in that their functionality is limited, whereas, on the other hand, their possible use in practice is already too difficult because of the technical expertise required in their operation. Therefore, a distinction is made between different types of users of spatial information. These distinctions have important consequences in the area of practice, in particular in the application of the present generation of GIS software. The current possibilities of GIS usage and their strengths and weaknesses are discussed in relation to the requirements of these various users. A strategy designed to meet these different forms of use is outlined. An information system called RIA (Regional Information through Automation) has been developed for public decisionsmakers and involved citizens in the Netherlands. It is based on a large and flexible GIS. The context in which RIA has been developed is discussed and a technical description of the system provided.

The evolution of spatial and nonspatial information systems is rapidly altering the role of applied computing in urban and regional planning. Urban development tracking and monitoring systems are in place in many regions in the United States and Canada. These systems provide continual status reports on land-use activity. Also, spatial interaction models are being developed for several major cities in the United States. The corollary collection of many types of spatial information has been impressive, The US Census will produce TIGER geo-referenced files with the 1990 census, and information will be widely available on CD-ROM disks. In this paper current trends in the development of new data sources, information systems, and spatial modelling are discussed and the potential for integrating these elements into a working planning-support system is discussed.

In this paper it is proposed that relevant information systems which support the activities of planning and decisionmaking must be based upon a thorough and clear analysis of the planning processes adopted. To this end, a general schema which identifies four key elements of a planning problem—opportunities, policies, options, and decisions—is introduced and various alternative planning processes are shown to be particular combinations of these elements and the links between them. Information relevant to the planning problem is influenced by the uncertainty pertaining to factual data and to knowledge concerning the problem in hand. Putting together the elements of the planning process with these various types of information leads to a threefold hierarchy of information systems: those at the bottom level concerned with source-related information; those based on intermediate-level routinised operation, such as diagnostic and scanning systems; and those at the top level which incorporate models and high-level intelligence in the form of decision support. An important implication of this paper is that for the successful design and application of information systems, planners should examine the wider context of their problem, process, and context in which planning takes place before selecting and developing such systems.

Powerful computing tools, including geographic information systems (GIS), computer-aided design (CAD), and thematic mapping packages, have been developed for producing maps and manipulating geographic information. Various types of computer-assisted mapping systems have been incorporated into planning practice. Research has shown that a significant number of local planning agencies in California either had or were planning to implement some type of mapping system. The experience of agencies that have already developed automated mapping systems can greatly benefit those just beginning to move into this new technology. A mail survey of Californian planning agencies shows that they use a large number of minicomputer systems and stand-alone microcomputers for mapping purposes, that a majority of the systems are some type of GIS, and that most of the thematic mapping packages are in agencies with larger systems. The major software packages being used are identified and an attempt is made to see how hardware and software configurations are changing over time. Land parcel mapping and traditional planning tasks related to the general plan appear to be the most important urban planning applications of these existing systems. Implementation and development problems based on hardware, software, and institutional factors are also explored. The respondents were either pleased with their investment in this technology or thought it was too early to tell. A few had experienced some cost overruns, generally in the area of base map preparation of attribute data acquisition.

Design criteria for urban and regional information systems are outlined. An attempt is made to evaluate the nationally available statistical series in the United Kingdom in the context of those design criteria and some of the systems developed in British local government to counter the deficiencies of an inadequate national data system are reviewed, Various systems for collecting local data have been instituted by particular local authorities, although the problems of using such data are often as problematic as the lack of those data in the first place. However, the New Town of Telford (Shropshire, England) has pioneered various local data-collection methods and an integrated information system which has been designed to monitor the demographic, economic, and social development has been built. This system is described, and it is suggested that, before such systems are generally possible, the quality and availability of data as well as their relevant usage within the planning process must be addressed.

Geographic information systems (GIS) are designed as general-purpose systems to manage spatial information. They are similar to commercial database management systems except for their strong spatial context. This recent technology has inspired many users who have, for generations, been faced with the difficult task of making decisions based on spatial information. The development, design, and functions of GIS are reviewed. Then effectiveness of contemporary GIS in managing spatial information in an urban environment is evaluated. In particular, special problems caused by high data density and multilayer organisation are examined. A solution to these problems demands special display, database query, and data-organisation components of a GIS. An application using CARIS (Computer Aided Resource Information System), a GIS developed in Fredericton, New Brunswick, is presented. This application illustrates the concept of a multilayered GIS with which databases of progressive detail can be vertically linked to facilitate data query.

This is the third in a series of three papers in which the methodological status of Q-analysis is investigated. The methodological rules abstracted in the previous papers are brought together, and the possibility of them being built into an expert system is discussed. A preliminary specification for an expert Q-analysis system is given. It is concluded that Q-analysis is indeed sufficiently replicable and procedural to be called a methodology. This completes the thought experiment set up in the first paper of the series.