5 replies to this topic

[Laura Miles]

Hello GIS experts,
Our organization has received logs of GPS points from units on trucks for a time period of a month. Right now we're trying to brainstorm with our transportation planners on what we can do with this data. Some of the fields included are time (to the millisecond), speed, and some other columns we are awaiting clarification on. I just wanted to put it out there to see if anyone had some interesting suggestions for what we (a government organization in the transportation industry) could do with this dataset. There are over 3 million data points and it feels like we're sitting on a goldmine that we just don't know what to do with!
Thanks in advance for your suggestions,
Laura

[frax]

Hi - aren't you going about this the wrong way - you should start with the questions that you would like to have answered using this data... ?

[Dragons8mycat]

I have been analysing Ship Vessel Traffic for navigation risk assessment and if may be worth you having a look at www.marinetraffic.com to see how they present "live" vessel movements.
ESRI have some good solutions worth looking at too, I think you would do well to read over the "Network Analyst" solution they sell, maybe not to purchase but to see what can be done.

@Dragons8mycat

[Laura Miles]

Frax - it's true, we're sort of going backwards. We've told our planners exactly that, come up with some questions you want answered and we'll see what we can do. I was hoping someone could give examples of the sort of things they may have done with similar data. The data we've received is sort of a by-product of another department's project, so nobody was expecting it.

Dragons8 - that's pretty cool, we have a similar map on our website. We don't get live data for the trucks though, and there is such a huge quantity it wouldn't be practical to show in this way.

[mfarmer]

At the least you can empirically identify or confirm truck routes, chokepoints, drive times, overlay with crash data, validate truck/passenger car ratios for your traffic modelers. You'd have to work with planners to examine how the data work with your regulatory framework. For instance, look at stops longer than say 10 minutes and compare those to your land use data. Retail? Industrial? Warehouse? Vacant? Etc.

Do one or two of those analyses before you meet with the planners again to whet their appetite. Make sure to invite the one who's you biggest ongoing PITA to that meeting.

So, how did you come across these data?

[Laura Miles]

Thanks mfarmer...some of these ideas we've thought of, others are new.
Our organization did hire a consultant to produce this data. It's a bit weird as we have someone doing GIS in one department, and then us doing GIS in another department, and there's absolutely no coordination between departments. (Bureaucracy at its finest!) So we were not consulted on what would be of value, or even notified that this data was to be produced, but somehow a transportation planner got his hands on it and shared with us. We're thinking about how to determine origin/destination, and thus trip times. I've already produced some heatmaps overlaid with land use, and a time lapse video of a day's worth of data which the planners liked. I think you're right identifying stops (as opposed to just gps logs) overlaid with land use would be more valuable.

Any other ideas anyone wants to throw out there are welcome!