3 signs your e-commerce logistics are obsolete - Part 3

3 signs your e-commerce logistics are obsolete – Part 3

This is the third and final parts on "3 signs your e-commerce logistics is obsolete" about the importance of route planning optimization in the current context of rising home deliveries.

29.09.2021 5

3 signs your e-commerce logistics are obsolete – Part 3

This is the last of three parts on “3 signs your e-commerce logistics is obsolete” about the importance of route planning optimization in the current context of rising home deliveries. You may wish to read the first and second parts before if you haven’t done so.

Sign 3: Each driver is assigned to a zone

Let’s recover the example of Acme, a grocery company selling through their e-commerce website to customers sitting at home.

Customer checkout their orders at the e-commerce website. Then it comes the time to close all bookings and start delivering.
If you have ever planned routes you most likely found yourself looking at deliveries displayed as markers spread across a map. A lot of markers.

Immediately you realise that in order to come up with a route plan going through all of those markers you’ll have to divide and conquer. This is the rule-of-thumb we usually follow to solve large and complex problems. When faced with a large problem we tend to split it into several smaller problems which can be solved individually more easily.

So we split the map into several smaller regions. To each region we assign a driver. Then we route all deliveries within each region to the corresponding driver. Hopefully we established regions whose average workload is balanced evenly across all drivers.
With time drivers get used to their assigned regions and get more and more efficient at finding and delivering to their assigned customers.

While this approach works it comes with a price.

Establishing hard regions may be too rigid under some circumstances. Consider the following scenario where one delivery is on the edge of a driver’s territory (in blue) and very close to the deliveries of the next driver’s territory (in red).

The driver in the blue territory will do 3 deliveries on the North-West region of his territory and then will have to come down all the way South to deliver to the fourth customer.

In the meanwhile, the driver in the red territory will have 2 deliveries close to the border with the blue territory. And he could easily do that extra visit. But he won’t because it’s on the territory of his fellow driver.

Having the driver on the blue territory to stretch his trip from the North-West to the South just for one delivery in very inefficient. But if you split into individual regions to simplify the resolution of a large problem then you lose visibility across regions.

Splitting territories is a good approach when you have to use such technique but sometimes counter-productive. Dropping such territories completely may not be an option because when drivers get used to their own regions their productivity increases significantly.

The state-of-the-art approach for this circumstance is to allow overlapping territories. While maintaining the advantages of established territories, this approach dissipates the issue with hard borders.

In fact, deliveries close to territories’ borders can now be served by any of the drivers, as in the following picture.

Both red and blue territories were stretched by one single block. It was enough to create the shadowed intersection area, but it could be a wider or narrower intersection.

Routyn supports several modes of territory splitting. First of all it supports no splitting of territories at all. If you don’t believe your drivers will benefit from driving always around the same region then there is no advantage in imposing such limitation at all. This means standard dynamic routing.

However, many logistics operations benefit from grounding drivers to the same regions every day. For those Routyn allows establishing non-overlapping territories but it also supports very well the added complexity of overlapping territories. The decision is always based on a cost-benefit analysis considering all variables and viewpoints.

Conclusion

The road to e-commerce logistics excellence is paved with hurdles. Most of these come from obsolete ways of planning routes back when systems like Routyn were not available.

However, new state-of-the-art procedures rise with the advent of advanced algorithms in the fields of Mathematics and Artificial Intelligence.

These are the roads being pursued by the innovative industry leaders on e-commerce logistics.