Enterprise GIS approach for urban transportation planning

Enterprise GIS approach for urban transportation planning

SHARE

Balamohan N, SP eResources Pte Ltd.
Balamohan N
SP eResources Pte Ltd, Singapore
[email protected]

Introduction
Over the last decade planning authorities in developing countries moved from conventional approach to integrated approach in urban planning activities. Geographic Information System (GIS) has become most promising technology for the integrated approach. In spite of the technology, the terms ‘haphazard urban growth’, ‘irregular pattern’, ‘rapid urbanisation’ have become very common among urban planners. This is because, lack of reliable information flow on time during the planning stage both in evaluating current transportation projects and proposed solutions.

Though the GIS technology has made planning easy and reduces the complexity in the traditional planning process. The problems are due to sequential impact of the implementation of GIS technology in project level rather comprehensive level.

This paper attempts to present the implementation of enterprise GIS concept in the area of urban transportation planning. This paper covers two major topics. First the concept and components of enterprise GIS with the real life example adopted in State of Qatar. Secondly it explains the linkage between the supply and demand side components of transportation planning with GIS. The overview of interfaces and analysis tools developed using the Qatar’s enterprise GIS Model has also presented in this paper.

Enterprise GIS
The concept of enterprise GIS is to maximise the utilisation of GIS capabilities by channelising timely spatial and non-spatial data from various source with in a government/organisational setup.

In the beginning of 1989, State of Qatar identified the potential for enterprise GIS, and now it is one of the most successful examples that implements enterprise GIS at national level. The Qatar enterprise GIS Model focuses on the common GIS requirements of spatial, non-spatial data and applications for inter/intra government agencies. The model determined that the following are the key components for the implementation of enterprise GIS;

  • Organization
  • Familiarisation
  • Prototyping
  • Standardisation
  • Cooperation
  • Base mapping
  • Data sharing
  • Applications

An administrative model has also been developed for the accountability of these key components viz., GIS Steering Committee and the Center for GIS. Taking the advantage of these key components national level transportation planning model was developed using data capturing and management capability of ArcInfo and modeling capabilities of the emme/2.

Role of ArcInfo and emme/2
Network and Land data can be easily developed, maintained and updated in GIS database. Transportation Planning and management needs accurate and timely spatial and non-spatial information like, network, capacity, speed restriction etc., to assist planning activities. Most importantly GIS database maintains and provides topological relationship (connectivity and contiguity), which plays key role either in macro or micro level transportation planning analysis. Standardisation and Data Sharing are the two components provided strong support for implementing the enterprise GIS in urban transportation planning. Arc/Info and emme/2 are the two system employed respectively for GIS and transportation planning/modeling.

ArcInfo is the comprehensive GIS suite capable to create, manage, analyse and present the geographic and attribute information in a more systematic and consistent way. emme/2 is a state-of-art graphical software tool for multimodal transportation planning. This system helps transportation planners, to model transportation network and assign the generated traffic under given set of conditions. There is close relationship in terms of data model between Arc/info and emme/2 has been identified for supply and demand side components of transportation planning.

In both the packages, the network model constitutes with series of nodes and links, however different terminology is used. These node and links are complemented with set of system-defined and user-defined fields. Both network models adopt network topology e.g., there is no need to create additional node at the intersection of a road and flyover. However one major difference between the data model is that the links in Arc/Info by default are in bi-directional, but in emme/2 each link represents uni-direction along the direction of link e.g. a two-way street is represented with two arcs in emme/2. The demand side information is linked with a spatial entity called traffic analysis zone (TAZ). The TAZ is maintained in polygon feature class in the GIS environment.


Fig 1: illustrates the concept of data sharing adopted in the Qatar’s enterprise GIS Approach for transportation planning and modelling

Supply Side – Network Extraction Tool
Basically transportation network and road inventories are the major supply side information for transportation planning and modeling. The following GIS functions are identified as more critical for the supply side data model to carry out traffic assignment / simulation:

  • Linear network information (representation of roads in relation with spatial coordinates system)
  • Directional information (one way streets, turn restriction)
  • Spatial measurement along the directional links (length of road segment)
  • Topological relationship (connectivity)
  • Representation of the feature (representation of curved road link)

As representation of directionality between Arc/info and emme/2 is different, the road network data model has built with an additional attribute to identify non-existent links. The non-existing links in Arc/info data model are flagged by setting the reverse direction (item TYPE_TF) with 99. During the export process, the interface ignored such links to port into emme/2 model. This data model also helps to enable or disable particular segment of road(s) for purpose of creating scenarios e.g. evaluating introduction of one-way street in the GIS environment.

The interface developed to extract supply side components for transportation planning has four modules. The modules are:

  • Editing (for Network, junctions, turn-restrictions)
  • Network Check (for checking the completeness of the attribute data)
  • Project Development (formulating scenario for decision making)
  • Export (Network and Turn Restriction)

Arc Macro Language (AML) and certain level with AWK and PERL scripts are used for the development under Unix environment. Roads department maintains the data accessed by this interface.

This interface provides two data files formatted for batch inputting in the emme/2 module 2.11 (network batch input)
Demand Side – Trip End Extraction Tool
Generated trips are the demand side entity for transportation planning. Land Use and Population are the dominant variable in the demand model. Travel demand is estimated for each TAZ. The source of data for land use, TAZ and population are maintained by different agencies. There are four look up tables derived for trip generation land uses, travel pattern, trip rates for each land use and populations.

Three basic modules are developed for:

  • Extract trips ends for each TAZ
  • Extract area for each TAZ
  • Editing Tool

The trip ends for each TAZ is derived based on multiple linear regression model. The interface delivers two files for trip production and trip attraction in origin matrix (moxx1 ) and destination matrix (mdxx1 ) formats. (xx refer to matr ix numbers) These files are used as direct input to the emme/2 module 3.11. The land use areas are extracted with reference to each TAZ for secondary analysis.

In both the cases (trip ends and areas) the extraction of data can be even restricted to only one TAZ. This option improves the performance of the system. The editing tool is basically used to change land use code to test the preset scenarios e.g. evaluating impact of change in land use.

Desire Line Diagram
Using Enterprise GIS model, a tool is developed to analyse travel characteristics and generate desire line diagram. The desire lines diagram is schematic representation of travel desire between origin and destination. The diagram is represented with straight line connecting the trip origin and destination zones. Relative widths of the lines indicate the relative amount of travel desire between the zones. This tool generates thematic map of desire line with population information. Relative sizes of the circles indicate relative trip generating or attracting power of the zone represented in terms of total population.

In order to generate the desire line analysis maps, three data sets are used from the enterprise GIS database viz., population, travel demand and administrative boundaries. Population and administrative boundaries are maintained by Planning Council (formerly Census Statistical Organisation) and Planning Department respectively. Travel demand is derived from the emme/2 transportation-planning model.

Generally the travel demand is calibrated at TAZ level and maintained in emme/2 demand matrix format. Using emme2 matrix editor the travel demand is aggregated into desire spatial level (say municipality level in this case). The aggregated demand matrix is punched from the emme/2 databank. This aggregated demand file is used as an input to generate OD_Table in the INFO. In the OD_Table demand points are coded with municipality ids. SPIDER command is used with administrative boundary coverage and O-D Table in Arcplot module to generate OD line.

Similarly the population information is aggregated at municipal level and available as INFO Table. The INFO table used as lookup table with administrative boundary. The size of the circle is calculated relative the population data and defined by SPOTSIZE command.

TAZ Validation
TAZ is a basic geographic unit used for transportation planning and modeling. Temporal changes in the geographic extent of the TAZ with respect to trip generating variables have sequential impact in the travel demand forecasting. During 1994 initial TAZs for were derived from census blocks, which was delineated during 1987 census. Due to rapid change in the development and implementation of various sub-division projects, the census blocks are changed and re-delineated during 1997. TAZ validation task was undertaken to identify the consistency of the TAZs against the new census blocks. The data (land use, cadastral and census) used for the validation process are sourced from three different agencies. This task is carried out using multiple map overlay technique in Arc Info environment.

There are two basic criteria has been framed for identifying the errors in TAZs and as follows:

  • TAZ should not bi-sect or cross the existing and/or proposed sub-division
  • Homogeneity in terms of land use distribution

This process clearly identified that the TAZs exists around the sub urban of Doha area are greatly affected.

Benefits
Traditionally in transportation planning and modeling the tasks are carried out with a defined set of assumption on the currency of data used for analysis. The transportation planning and modeling team enjoyed the strength of enterprise GIS in Qatar during the development of national level transportation planning model. Some of the major benefits are:

  • Received timely spatial and non-spatial data during development cycle of the model
  • Minimised (in fact eliminated) lose of information during the data collection
  • Conflicts of source and redundancy of data is minimised
  • Frequency of data capturing and updating for the respective database is maximised.
  • Minimised misestimating of trip generation, as the land use and population data are readily available via WAN during continuous case by case transportation impact studies due to land use change.
  • Presented the transportation analysis reports in a more meaningful way with the help of supportive information available with another agencies.

Enterprise GIS Approach for India
Enterprise GIS approach has been proved most successful in State of Qatar. Implementation of the same concept in developing countries like India seems to be a myth. It is because of the size of the country. However, if we consider individual components of the approach and scale of implementation it would become a reality. The National Spatial Data Infrastructure (NSDI) Bill is the right example and it is the initial milestone in the Indian GIS industry’s approach towards enterprise GIS.

It could be more rational to consider the scale for initial deployment at metropolitan level. Some of the advantages for this consideration are;

  • Urban planning problems are more in metropolitan areas
  • Well defined area for implementation
  • Short term development plans become more important than the long term master plans and thus planning authorities require timely flow of most reliable information
  • IT infrastructure is readly available
  • Agencies like Planning & Development Authority, Highway, Utilities & Infrastructure, Universities and Research institute etc., are centrally located

Conclusion
Qatar’s experience provides a model for governments and organisations in developing countries for implementing enterprise GIS, e.g. Jamaica, Vietnam. Organisational framework, standardisation, cooperation and data sharing are the more critical component of enterprise GIS. This paper also concludes that the fundamental aim of transportation planning and management i.e., providing safe, efficient and affordable transportation system to the society, can be effectively achieved with the implementation of enterprise GIS.

Disclaimer
The view expressed in this paper are only the author’s experience and with reference to other published sources and do not echo the governments, organisations or firms by whom the author has been retained.

Reference

  • Balamohan N, Dr.Rashid-Al-Matwi, 1999, Travel Desire between Municipalities and Doha, ESRI Map Book Vol. 14, pp 109, ESRI, USA.
  • Balamohan N, 1998, Data Exchange Protocol for Land Use Transportation Planning, Al-Khabar Vol.3, No.1, The Center for GIS, State of Qatar.
  • Jerry C. Coiner, 1997, Transferability of the Qatar Enterprise GIS Model: Experience in Vietnam and Jamaica, GIS/GPS Conference’97, March 2-4, 1997, Doha, State of Qatar.