Special Projects - MAIN

Introduction | Purpose | Maps | Scope | Finding the Data | Summary

Where is the Data?

Identifying Component Resources

Data resources necessary to conduct the project were not easy to come by because no single government entity within the state was responsible for maintaining location-based information on all the others. 

One governmental unit might have very good location data for each of the hospitals serving the state, but have no information about the locations of schools. This introduced quite a challenge just in collecting a baseline of information to begin connecting government services with a location. 

However, by taking advantage of geographic information technology, an asset management system could be built for governmental resources.

For the purposes of this project, it was determined that a network consumer would be any organization using the state network for delivery of government services or healthcare. This covered a broad scale of data sets, which presented one of many challenges. It also covered all corners of the state near and far, another significant challenge. 

Two primary categories were chosen as relation points within the project framework:

Existing and Potential Network Consumer Categories

Existing Network Consumers (ENCs)

These are organizations utilizing the existing network services provided by the Arkansas Department of Information Systems. By visualizing their locations, systems and services could be better managed.

Potential Network Consumers (PNCs)

These are potential organizations which could utilize network services provided by the Department of Information Systems. By visualizing their locations relative to current service points, opportunities for additional connections could be identified and managed. This would also provide opportunities for pooling consumers at a single point of presence.

• Education: Public Schools, Private Schools, Libraries, Museums, Universities

• Criminal Justice System: Jails, Courts, State Police, County Sheriff, City Police

• Health Care: Area Health Clinics, Hospitals

• Social Services: Employment Security Division Offices, Department of Human Services Offices, Community centers

• Other Government Services: Fire Departments, City Hall, Game and Fish Commission, Revenue Offices, Military Installations

Five Essential Facets to the Project

• Capture location data for all ENCs

• Capture location data for all PNCs

• Organize all ENCs, PNCs and the Network into one geographic system

• Design an online interface to the system

• Serve the system online through a browser

The ENCs were stored in a billing database at the Department of Information Systems. The database was used exclusively to support the billing process and was not readily usable as an asset management tool. The billing database did not have a geographic component other than mailing address information. The project team determined the billing database represented the best single source of information about the ENCs supported through the network. Formatting issues represented the only obstacle to using the table in an address geocoding engine to address match and geocode the database. 

The Telecommunications Engineering staff at the agency developed a mapable design that would support geocoding and also future attribute queries; for example, a query for all Department of Health offices that have T1 access, etc. After the table design was completed, the entire billing database was processed through the address geocoding engine to generate coordinates for the data. The relative location of all the circuits could then be displayed. During the geocoding process, a fallback to zip code centroid option was selected, so for records with no street address match a coordinate was placed on the zip centroid. Over 2,900 records could be mapped and displayed for the entire state showing ENCs locations. For the first time policy makers and executive management could visualize where resources had been allocated and also where other areas were underserved.

This portion of the project was accomplished with relative ease; however, the coordinate locations of the ENCs were not reliable because of the geocoding process. For records that could not be geocoded to a street address location, the coordinate placed on the zip centroid meant location data could be incorrect by several miles. The team needed a way to validate and refine the ENCs coordinates across the state.

The PNC data prior to this point was not stored anywhere centrally, nor organized in any mapable fashion. Although an agency director might have precise information about the location of the assets under management, they would not have location data about all government assets within the state. Capturing all the unorganized data represented a significant exercise in address data collection and organization. For example, a rural water association would not necessarily be managed by a state agency and its location not stored any single place. 

However, as the e-government wave builds within Arkansas, it may be possible to furnish the water association the capability to report monitoring information back to the regulatory agencies such as the Department of Health. The reporting could be handled through e-mail, a web browser, or other network interface. However, in order to engineer this connectivity, the network designers must account for its location in relation to existing infrastructure. These and many other potential users needed to be identified and located. After the data collection task was accomplished, this data was also parsed through the address gecoding engine to provide coordinate locations.

Introduction | Purpose | Maps | Scope | Finding the Data | Summary