What Don't Shipping Companies Have Better Delivery Estimates? 2012-07-12 Disclaimer ---------- 1. This post is based on my last experience with UPS, but I am sure that it applies to other shipping/delivery companies as well. 2. I really like UPS and will continue to use them - this post is not intended as a bash on an otherwise decent company. 3. I don't know what kind of logistical data UPS keeps/has. Below follows from only my own suppositions. The data is already there ------------------------- It is probably safe to assume that UPS knows exactly which packages are on which truck, the route that each truck is to take, and expected route start and end times. It is also probably safe to assume that the current location of each truck is easily discernable. If each truck is not explicitly tracked by GPS, then by keeping track of when each package is delivered a reasonable location can be inferred. Combine these data with traffic data from Google and it would be fairly straight forward to give a very precise delivery time estimate. So why don't they do it already? -------------------------------- I can see already that this might be a gray area in the Google Maps API ToS. I see in them that there is a distinction of Google Maps API for Business and the normal API. From what my cursory read-through uncovered, since I can track any UPS package without needing to login or be behind a paywall, it would seem that the normal API could be used. If I were making this service, I would still have a lawyer check out my reasoning. So maybe the cost of realtime traffic data is prohibitive. The service that I am proposing is still technologically viable because traffic data was just icing on the cake - without it, a reasonable 10min delivery window could still be generated. Also, just the sheer volume of packages handled by UPS each day might make this service very resource heavy. I will argue though that if UPS can handle those logistics internally, it shouldn't be too difficult to present those logistics externally. Or maybe there isn't enough monetary value in this to make it worthwhile. Wouldn't this be an awesome project to work on? ----------------------------------------------- I think that this would be an awesome project to work on. Every single person that ships a package sits there at their computer clicking and clicking the refresh button on the tracking page. If this service was implemented, they could go the the tracking site once and see that their package would arrive "between 9:30AM - 9:40AM Tomorrow." Even from a technological view this project would be amazing - Real-time data analysis and visualization. A public facing API could be built and then there could be smartphone apps that alert you when the truck that your package is on is at your house (or inadvertently delayed) so you could rest easy knowing that your package is on its way. Closing Note: ------------- I would drop everything to work on this project - I mean that. If there is a shipping/delivery company that wants to build something like this, I would absolutely love to work on it. I could see this service being a good threshold for a consumer shipping company to get into time sensitive deliveries. I could also see this being a semi-passive income site that acts as a third-party package tracking service.