Nobody Walks in L.A.
For those of you living in congested metropolitan areas, such as Los Angeles, you understand the true meaning of traffic, where we measure travel to destinations in terms of time, not distance. We now have GPS systems suggesting optimal routes and re-routes with time estimates to arrive at your destination. The amusing part is jumping on the freeway at 3 pm and the GPS says 59 minutes to arrive home. Then, 15 minutes later, it still says 59 minutes to arrive home, or it may announce that there’s been a delay on my current route and would I like to take an alternate route. Similar scenarios can repeat multiple times, and the actual time to arrive home may actually be 90 or 119 minutes. So, were the GPS algorithms off?
The GPS is analyzing potential routes from my current location to home and traffic as it is right now. It’s amazing to have that bird’s eye view based on the analysis of thousands of roads and current conditions. What missing is the more general, human knowledge that, hey, it’s rush hour and cars will start loading up on the streets and freeways at a rapid pace starting at around 3 pm and continuing on through 6 or 7 pm, and traffic will become increasingly congested, also increasing the chance of accidents and breakdowns, or that there’s a Lakers game at the Staples Center downtown causing major jams rippling outward.
The seemingly neurotic behavior of the GPS is because it is only looking at traffic conditions in present time and reacting as new situations develop. It is not tempered by historical knowledge of typical traffic patterns for that time of day during the week on those particular roads or that there’s a ballgame planned for 7 pm – all the things we used to think about when planning our routes before GPS came standard in every vehicle and smartphone. My common sense based on experience tells me that if I leave Gardena at 3 pm, I will probably arrive home between 4:30 and 5, yet GPS will first show my arrival home as 3:59. I will even blow through some of the suggested re-routes as again, because it is only considering traffic as it appears right now, it is not taking into consideration possible changes and that staying on the original common route may end up taking the same amount of time or even less than the crazy alternates. An accident on my route 20 miles away could easily be cleared up before I even enter the affected zone.
Planning ERP or software implementation projects and resources can be very analogous to our modern day GPS. If you only base the projections of your project on the resources, other jobs, and business as it looks today, then you will tend to find yourself in a constant neurotic mode of reacting and scrambling to manage change. If you take a longer, broader view based on experience with similar projects, knowledge that if business is growing, then traffic may increase, putting heavier demand on existing resources, not to mention increased incidents – stalls, jams, bottlenecks, resources leaving, vacations, backlogs, two projects demanding the same resource – then you may be able to forecast with a little more reality and accuracy. No one has a crystal ball, but it does answer the question, why does a 400-hour project with 4-5 resources take 5-6 months to complete? And there are often multiple parties involved.
Prediction and forecasting are still possible in these scenarios if we recognize that we are dealing with many more factors than just a mechanical look at the most efficient route from Point A to Point B without taking into consideration a multitude of changing conditions that can affect the outcome of our journey. Maybe someday GPS systems will have better AI, but for now we must apply real reason to achieve more realistic outcomes.