PROBLEMS AND ERROR 

With any project where data has to be sought out to be used for a particular software systems many problems exist. Using IDRISI 32, your data should be in raster format. Unfortunately, there is not a vast amount of IDRISI data readily available for download off the web. To answer my spatial question, I had to go to ESRI to find a data set which was applicable. This data was in ShapeFile format. Using a Vector data set in a raster based program, points and lines are generalized and attributes are lost in th conversion. Much of my data aquisition problems arose out of having to manipulate and link attribute files to the now raster data. To create my least cost pathways I had to digitize the firehalls and hospitals so that they would fall directly on the roads layer, in some cases slightly altering thier position.

Using database workshop I had to manually filter out each street and assign it a value to create the roads layer. I used a New York City road map for this process which led to incorporation of my own biases based on my perception of the map. What one person may have assigned a Major route, I may have assigned a Secondarily Major route. The speeds I used to create my friction layers were general estimations, as hard data would be next to impossible to obtain unless you were in the traffic yourself. The friction layers also did not incorporate intersections, where two roads meet. When an emergency response vehicle would have to greatly drop their speed to cross an intersection. Given more time it would have been interesting to investigate this property.

Although there were no collosal problems associtated with my methodology, and I didn't have to change the direction of my analysis, small error is going to accumulate throughout the process. My analysis provides a good general overview of emergency response times and routes, but would have to be finely tuned to be used for a real life crisis.