Sometimes, close enough…is just NOT CLOSE ENOUGH. When it comes to developers building location-aware apps, more often than not, they decide to use formulas that compromise accuracy for quicker results.
Now while quicker results are just fine for most simple applications, there are times when you need the highest degree of geo-spatial accuracy in order to provide pinpoint results for the user. The funny part is, mobile developers in general are missing this critical formulaic implementation, as it pertains to enterprise level economic calculations. WTH am I talking about, you ask? Well, let me explain.
Let’s say you have a client in the OTR Trucking business, who needs to calculate fuel prices based on location, distances between refueling stations, load destinations, weigh stations, toll-roads, detours, etc…all on the fly and in as precise a measure as possible? What if I told you that many transport companies have learned the art of, “rolling into weigh stations on fumes”, in order to save significant money and delivery delays based on weight? This practice is quite common in fact, and a credit to the OTR folks knowing how, when, and where to save at a maximum level. Since savings equal more profit, you can’t help but be impressed with how trucking companies and truckers in general, have cleverly overcome the significant obstacles in the way of them bringing us our daily goods.
So what do these things have to do with “precise” GeoLocational strategies? And does it really matter that much? Answer: YES!
If we go back to our previous questions, let’s suppose a trucking company in Nebraska has a major load to be taken to Washington D.C.. With every twist and turn aside, assuming we could drive straight to D.C. as the crow flys, and for the sake of argument, supposing our headquarters is calculating costs using the three GeoLocational formulas known: Haversine, Spherical Law of Cosines, and Vincenty Formula, what are the results? Let’s take a mathematical peek.
Using the most popular GeoLocational formulas in the mobile development space, Haversine and the Spherical Law of Cosines (vs. Vincenty), we would end up missing our D.C. destination by as much as 53km (or 33 miles). If we had calculated location via Vincenty, we would have nailed the heart of our delivery location in Washington down to 5mm!! So regardless of how bad your math is, 33 miles off versus 5mm, is quite significant.
Moreover, this example was based on the idea that we would get to Washington D.C. “as the crow flys”, which we certainly cannot. Therefore, calculations using the most popular GeoLocational formulas (Haversine & Law of Cosines), would put us off even more!
At the end of the day, what does knowing this actually do for us anyway? Maybe more to the point would be, what would I do (as a developer) knowing this information? Well it’s quite simple to me, I would build two products that “spoke” to each other: a desktop app that pushes data to a mobile app.
The desktop app would process the distance equations, calculate stops, turns, weigh stations, toll roads, weather and traffic congestion, and fuel prices in real time. The desktop app would then “push” that live data out to the mobile app, and the mobile app could sync itself in real time with the data coming from the desktop portal. Working in concert, the desktop app could perform the heavy lifting by processing massive amounts of real time data calculations and running formulas, while the mobile app sat in wait for the “approved & certified” notifications to come its way. When it’s all said and done, the guesswork is eliminated for the driver, money and time are saved in multitude for the business (which equals greater profit), and operating efficiency is not only perfected, but sustained!
All because a smart developer paid attention to the importance of hyper GeoLocational accuracy! Now imagine that?