GISdevelopment.net ---> GITA 1998 ---> Integration of the Enterprise

GIS Systems/Data Integration Success Guide

Ricardo Salazar
GIS Project Manager
Metropolitan Water District of Southern California
350 South Grand Avenue
Los Angeles, California 90071


Abstract
Can a company successfully integrate diversified GIS Systems and Data? The first step to successfid GIS data and system integration, is to develop an awareness of potential problems in data integration; as data is perceived differently, and system integration, as hardware and software issues will dictate interfacing capabilities. Find out what potential problems may exist and how to identify them, so you can build a successful GIS project.

Introduction
When developing a GIS project, one must apply his previous experience managing projects in other sectors, many of the issues found in those projects relative to financing, staffing, training, systems, etc. will also be valid for a GIS project. This paper focuses on problems and possible solutions integrating data and systems, in the process of developing the base map and applications of a GIS covering a large geographic area.

Systems/Data integration merits detailed study for two main reasons, the first because this is a new area for many project managers, and the second one is because those areas now include different elements that didn’t exist in Engineering, Surveying, Planning, Information Systems or other disciplines applications.

A company beginning to implement GIS projects normally assigns Project Managers with experience in one field or discipline closely related to the application that needs to be developed. GIS applications are multidisciplinary, normally needing to cross the boundaries of different disciplines and functional areas to implement an application. Another important issue is that implementation will require the use of other related technologies such as photogrammetry, cartography, Global Positioning Systems (GPS), remote sensing, Data Bases, Networks, Internet, CAD, HW (Server and clients), Surveying and Engineering SW, etc. A Project Manager with broad experience in the above will be more effective in managing the projects.

Other factors that we need to consider are the type of applications (base map vs. specific application), and size of the project since they will determine the complexity of the solution. For example the size of the project could cause a longer implementation time that could lapse two or more years, were changes in technology (HW, SW, Communications) can impact the implementation plan.

From personal experience, the main problems encountered, were to assemble the real data needs for the project, determine the key fields to be used across the enterprise, define the systems environment and the tasks and their sequence to develop the project.

The majority of the observations written in this paper are based in the present development of GIS at the Metropolitan Water District of Southern California. MWD provides approximately 60% of the water used by the nearly 16 million people living on the coastal plain of Southern California between Ventura County and the Mexican border covering about 5,200 square miles and including 240 cities and unincorporated areas. Several reservoirs, filtration plants and an almost 1,000 mile pipeline distribution system that includes the Colorado River Aqueduct are used to satisfy the yearly average of 2 million acre-foot water demand (one acre foot= 325,851 gallons).

GIS Development
Development of GIS applications began at MWD in 1990 and were used primarily as a planning tool within the Planning and Resources Division to plan MWD’S 3.9 billion capital improvement expansion program. More than 40 GIS layers have since been implemented to support this effort using ESRI’S Arc/Info and ArcView software.

In early 1995 the Engineering Division began to explore the use of GIS technology for its use, by developing an internal strategic plan and conducting various pilot projects using different SW packages. The driving force for the use of GIS technology in the Engineering Division and getting the approval of funds was a combination of the need to implement an application to pinpoint and manage high revenue generation locations, such as telecommunications sites and continuos Right-Of-Ways and to have an accurate location of the facilities and pipelines for MWD’S preventive maintenance and emergency response applications.

A key part of the project was forming partnership between the Planning and Engineering GIS Division functions, the Corporate Information Systems has also been very helpful in minimizing data and systems integration issues. Operations Division is now being added to the effort for data needs and standardization, this will allow us to complete the cycle needed integrate and manage infrastructure type of projects.

Data Integration
There are several questions related to data integration that we need to respond to and be aware of when implementing GI S in an organization:
  1. What data needs to be integrated and how will it be done?
    The first step to know what data will be used in an organization is to perform a GIS and Data Integration Strategic Plan, this plan will help to achieve several objectives:
  2. Identifj the fhnctions and tasks needed to implement GIS in the organization
  3. Establish short and long range goals and implementation sequence or schedule
  4. Identify the applications that can benefit from using GIS
  5. Identify the existing problems, their impact, solutions and benefits that can be obtained from using GIS technology
  6. Identify the needs and obstacles for implementing GIS (political, financial, organizational, personnel, etc.)
  7. Have a marketing and educational tool to promote GIS within the organization
  8. Identify the data used by key applications and their integration needs.


  9. What is the data status?
    After the applications that can be implemented are identified and prioritized, we will conduct an inventory of the existing data and compile the following information:
  10. Source of data, who generates the data, who does the updates and who uses it
  11. Existing volume (number of maps, records, documents, etc.)
  12. Medium (paper, scanned, vector, digital, film)
  13. Origin (where the data was created)
  14. Quality, reliability and accuracy
  15. What format (database, spreadsheet, word processor, mainframe)
  16. Frequency of use (query and update), importance factor, date created
  17. Alternative sources of information
  18. Data growth potential


  19. What are the data needs for the business functions, applications and data integration?
    Analyze what are the data needs and correlate them to the required business functions and applications. Examine any duplication in data from different finctions, evaluate the collection cost and alternatives for lower cost data.

    By installing GIS into the organization you have uncover data compatibility issues that are critical to the future of your organization. There are six major categories of data integration for Gee-Referenced information and if they are not addressed, the efficiency of automation in your organization will be substantially reduced. The si categories are listed below:
  20. Mapping coordinate system and datum such as State Plane, UTM, Lat - Len, etc.
  21. Mapping information spatial accuracy.
  22. Facility code systems or geographic identifiers for each objector location.
  23. RDBMS tabular data format. Ex: are names store as First, MI, Last or as one field.
  24. Mapping/GIS data file formats, such as AutoCAD, MicroStation, Arc/Info, etc.
  25. Map symbology.

    26.
  26. Are there any GIS mapping standards in the organization?
    From the time of the application design to the fill implementation of all the modules, could be a range from a few months to several years, during this time production of new maps and data will continue. From the analysis of the data select the frequently used data following the Pareto rule that gives us an approximation that 20% of the data is used in 80’%0of the functions. Develop GIS data and mapping standards for the frequently used data first and use the standards for creating new data in the existing processes.

    For example, if as part of your base map you will be using right of way maps, generate GIS standards to be added to the CAD standards to be used ASAP by the right of way function when drawing right of way maps. Following a sample of standards:
  27. The last coordinate pare of a digitized area boundary must exactly match the first pair.
  28. All polygons or area boundaries must be closed and tangent at each connection point with no segments overlapping.
  29. Roads, border line, property edges must be created using a single line string .
  30. All line feature segments tangent or intersecting with other line or area boundary features must be terminated at the point of intersection.
  31. Provide a list of the GIS items that will be used in the map and the levels to reside.


    32. From using this standards you can alleviate the conversion process. In an study conducted in MWD implementing the GIS standards for the right of way maps mean an additional work of three hours per map versus ten to thirty four hours of conversion time per map needed in the future.

  32. How does the conversion process take place?
    Decide how to integrate the data and mapping from legacy systems and projects that have been constructed with paper base information vs. actual or recently finished projects with digital or vector data.

    In previous steps we have found which are the data available, how it is used (query and updated), frequency etc. Based in this information we will decided which data needs to be converted and how (cogo, digitized), this will depend of the level of accuracy needed by the applications, cost and importance, normally the basemap layers need to go through this process. Other maps can be just scanned, geopositioning or indexed in a database and saved in CD rem’s and used as a reference. Don’t forget that sometimes the data can be acquire from other sources instead converting old information, for example parcel line work can be acquire from the Counties or Cities if it satisfies the level of accuracy needed by your application, this approach can save considerable amount of money to your company.

  33. How am I going to setup the data dictionary or Metadata function?
    The concept of data dictionary comes from 1980 when the usage of hierarchical first and then relational databases began to expand and the need of the organizations to keep track of the characteristics of their text data, usage and files or tables location was needed. The Metadata concept is to keep track of geospatial data, the content, quality, condition and other characteristics of the data. It will help to organize the data and provide information about your data to others within the organization as part of the integration processor externally to clearinghouses if desired. To perform this function you can develop your own database or use existing software like the California Geospatial Metadata Clearinghouse developed by the California Geographic Information Association.

  34. How the applications will be integrated?
    The applications will be geographically integrated by using a base map developed with the highest level of accuracy needed by applications using the same area. For example to manage infrastructure type of projects you go through a cycle of Planning, Design, Construction and Operations, each function has different applications and data accuracy needs, for the Planning function 100’ or 200’ base map accuracy will be sufficient for their applications and they can be integrated for that function. With Design, Construction or Operations applications, the accuracy needs are more stringent plus/minus 1 meter or less. The basemap for those areas or corridors has to be developed to that level and the rest of the areas needed by the Planning function can remain at their level of accuracy. In this way we can begin to use the same base map for all the functions and geographically integrate the applications.
Data Integration Summary
When using GIS the concept of data integration goes beyond previous systems that deal with text data or graphics, now by combining text data residing in one part of the organization, with maps, photographs, documents, etc. that can reside in other functions of the organization we get a synergy tool to visualize, analyze, manage an integrate the organization information.

A strategic plan will help us have a better idea of the applications in the organization, the data needs, standards implementation, and integration with non GIS applications by providing a geographic link component to the non GIS application using the corporate relational data base.

In other words we need to satisfy the organization’s information needs and plan and design the system to incorporate fhture informational needs of the organization. Good relations and communications, formal or informal committees with other groups of the organization will keep us aware of their respective applications and data needs.

Anticipating steps needed to integrate data in a GIS will help us succeed in the implementation and usage of the systems.

Systems Integration
To properly select the software and hardware that will be utilized for a project and to integrate it into the company’s information technology plan, the following steps can be followed to choose the optimal solution:
  1. Assess general technological trends or market direction in HW and SW.
  2. Assess current level of technology in your company and future migration path.
  3. Decide where you are going to be positioned in this technological environment.
  4. Based on the application requirements and previous steps, choose the GIS software and hardware to be used.
  5. Establish hardware, software, network standards and migration paths
  6. Applications development
Compiling this information and analyzing it, is a task that requires help from various sources, such as the company’s Information Technology department, other companies with similar applications, consultants, conferences, users groups, software vendors, publications, etc. Educate yourself in advance to better understand how to put the pieces together, be aware of the limitations of the technology, budget constraints, integration with existing and planned systems, security issues,. and personnel expertise. The following are some of the aspects of consideration relative to the above mentioned steps:
  1. Assess general technological trends in HW and SW
  2. What are the current operating systems in use and what are the future trends?.
  3. What are the types of communications, speed, cost, reliability, usability ?
  4. What is the expected development incomputer processor speed, video cards, plotter and printer technologies, remote sensing, GPS, aerophotogrammetry?
  5. What are the developments in databases techology adhowthey will help in the applications integration process, programming languages, CAD/CAE software, artificial intelligence, SCADA systems?
  6. What are the development trends of existing GIS packages and how do they comply with open standard technology for spatial data interoperability, connectivity with databases, OLE (Object Linking and Embedding), COM (Common Object Model), platform independent code, web enabling technology, 3D systems?
  7. What is the SW and HW mainstream market direction?


  8. Assess current level of technology in your company and future migration path.
    Extract information relative to the company’s migration plans, platform to be used, database engine, LAN operating Software such as Novell, Banyan Vines, TCP/IP (Transmission Control Protocol/Intemet Protocol), architecture to be used Client/Server or Thin Client will help in integrating GIS with other business functions and applications and concurrently integrate into IT’s standard systems support function.

  9. Decide where you are going to be positioned in this technological environment
    There are two issues that must be defined. The first issue is your company’s positioning with the latest technology (bleeding edge, leading edge, current stable technology). this will depend largely on the application needs, technology migration plans, cost benefit analysis, technical support availability and the level of risk that you and your team are willing to take.

    The second issue is your company’s plan to manage future technology changes to lessen the impact on your applications. Approximately 80°/0 of GIS implementation cost, go to data acquisition and format, the main effort will be to minimize the impact of changes in the access to data by storing it in the most common formats used by different vendors, relay in those portions of the system that age slowest like the databases, and plan to minimize the impact of updates for example keep the pipelines alignment information in databases instead of graphic format.

  10. Based on the application requirements and previous steps, choose the GIS software and hardware to be used
    Develop the criteria that you will use to evaluate the software packages and provide a weight to each of the items according to their importance, multiply it by the points assigned when doing the evaluation to get a weighted score. Here is some criteria that you can consider in your evaluation:
  11. Implementation time for: Database development, graphical user interface, training, testing and debugging.
  12. GIS functionality to match user requirements ofi Research, display/data analysis, thematic mapping, dynamic segmentation, topology & polygon overlays, 3D.
  13. Cost ofl HW, SW, training, development, maintenance, consultants.
  14. Availability of support: Inside the organization, vendor, consultants.
  15. Plotting and printing capabilities
  16. Easy of use: Display/query, customization
  17. Maintenance/Update: Graphics, database, applicationhser interface.
  18. Compatibility/Integration with: CAD, other applications within the organization, corporate IT development plan, usage of external data sources.
  19. Direction/Growth. Evaluation of the future products that the companies been evaluated are working on.


  20. Establish hardware, software, network standards and migration paths
    Even with the Windows or UNIX standards there are so many possible combinations of computers, operating systems, printers, interface cards, applications, device drivers, and networks that it is easy to create combinations that do not work or produce erratic results. When building out an application, target platforms should be tested first and only implemented when they prove their reliability. Generally most computers and software need to be completely replaced after a maximum of five years because the rapid obsolescence of technology. This means that the company needs to have a comprehensive IT strategy to migrate from legacy systems and for future migration needs.

  21. Applications development
    When developing the initial applications for large areas, the data/systems integration includes outside sources. For example, if you are acquiring the parcel line work from different counties, they will beat different accuracy levels, zones and coordinate system registration epoch which will need to be reconciled to integrate the information that also will came in different formats. Features in layers with a lower level of accuracy need to be adjusted and registered to the layer possessing higher accuracy. Scheduling software can be very helpful in integrating the tasks, sequence and resources needed to finish the project.
Conclusion
Knowing in advance which are the Systems/Data Integration issues that will need to be dealt with when developing a GIS project is the first step to a successful project. Educating ourselves and the organization as much as possible in the technologies and issues involved in the implementation process will allow us to better evaluate the proper use of the technology and resources for Systems/Data Integration.

© GISdevelopment.net. All rights reserved.