Intelligent Data Storage

All collected information, whether from field survey work, desktop mapping, or other input methods, is stored in a geodatabase that relates all collected information to objects and features in space. This means that information about a particular outfall is linked to geographic information, including the roadway, receiving stream, county, TxDOT district, land use, and drainage area. Each outfall, in turn, is linked to all of the inspection, hydrology, and desktop mapping results from that outfall.

According to Sonny Lelle, systems analyst/technical project leader for TxDOT's Information Systems Division, the heart of OTS is its approach to storing relational data. "With these linkages, powerful reporting and data visualization tools can be created," Lelle says.

In addition to an intelligent approach to storing relational data, OTS includes custom programming to process data and to populate some geodatabase fields automatically, without user intervention. One of these programs helps Texas storm water managers decide which outfalls require follow up investigations by placing each outfall into one of the following categories of illicit discharges: unlikely, potential, suspect, or obvious.

Artificial intelligence features of the OTS software classify outfalls by considering a combination of visual and physical observations of vegetation conditions, staining, or odors; field-measured water chemistry results; and laboratory-measured water chemistry results. The software program used to make the classification assignments runs overnight and identifies all new outfall inspection results loaded into the geodatabase from the previous day.

"Now storm water managers can review inspection results within a day to determine which outfalls require follow up actions," says Crisp. "That improves the effectiveness of our investigations or third-party notifications."