Connecting Route Planning and Execution for Profitable Wholesale Delivery

Too often, route plans look good enough on paper but unravel out in the field.

Some vehicles head out under capacity, while other routes are overstuffed. Driver overtime creeps up, and estimated arrival times (ETAs) are unreliable.

Wholesalers and distributors become resigned to these challenges over time. Delays and errors feel like inevitable realities of logistics.

But most of these issues can be avoided with more realistic route planning. The key is to integrate planning with execution to build feasible, profitable routes.

Here’s how distributors can move from theoretical plans to routes that work effectively in reality.

Why should wholesalers and distributors track route execution?

Today’s route optimisation algorithms can generate realistic, highly efficient routes. When routes go off track, it’s normally because the plan was based on inaccurate data and assumptions.

In wholesale markets like food, beverage and building supply, delivery performance relies on accurate data about:

• Precise drop-off locations (the dock, not the street entrance)
• Business opening hours
• Promised time windows
• Accurate service times – which can vary based on case count, product type and account type
• Access and vehicle constraints

If this data is “off”, routes become under- or over-optimistic, ETAs drift, and customer service suffers.

How does execution data build better plans?

The solution is to tie execution data back into route planning. Drivers’ mobile devices can capture actual behaviour such as drop-off locations and service times.

Then, Machine Learning (ML) tools can model how service duration varies based on product type or customer characteristics.

Route optimisation software uses this data to build feasible plans with realistic ETAs and predictable costs.

Key data for route planning

Wholesale distribution tends to have higher stakes than B2C delivery. One incorrect service time can throw off ETAs for all subsequent deliveries on the route. And late or failed deliveries have a knock-on effect on clients’ profits.

When it comes to delivery planning, these data points are particularly prone to cause failures:

• Business hours and time slots: Incorrect open/close times lead to missed and delayed deliveries.
• Service time predictions: One-size-fits-all values like “10 minutes per stop” don’t account for different SKUs, handling methods and customer types.
• Location and access details: Middle-of-the-street geocodes are a headache for drivers who need to find the precise dock location. Inaccurate geofences can also skew recorded service times.
• Driver and vehicle constraints: Route optimisation software should be able to consider drivers’ hours, breaks and skills. Equally, vehicle constraints, road restrictions and available capacity should factor into route planning.

Of all these areas, service time prediction has the biggest impact on delivery performance. Unrealistic service times lead to inaccurate ETAs, eroding customer trust. In turn, routes are under- or over-stuffed, leading to a higher cost per stop.

Closing the loop between execution and planning

We’ve seen how execution data is vital to keep wholesale delivery operations running smoothly. Here’s how technology can help you get started.

1. Evaluate current operations

Route planning software from Descartes highlights variations between planned and actual activity:

• Are you hitting agreed time windows?
• Are ETAs drifting from reality?
• Are drivers racking up unexpected miles and hours?
• Are drivers following the sequence?
• Is cost per stop unpredictable?

The software helps you identify whether differences are a result of driver behaviour or unrealistic planning.

2. Capture mobile execution data

Equip drivers with a mobile app that captures actual activity.

• True dock locations
• Accurate service times based on auto-arrive and auto-depart
• Proof of delivery with geo stamped and time stamped photographs.

3. Feed execution back into planning

Route execution software can detect when drop offs repeatedly take place outside of the expected geofence. The software can automatically update the stop’s geocode, or prompt a route planner to make the adjustment.

When it comes to service times, ML can model how the duration varies by weight, cases, volume and account type. This allows for more realistic capacity planning and ETAs.

ML tools can then predict ETAs with increasing precision as the driver moves along their route. This lets you share dynamic ETA predictions with customers, sales reps and dispatchers.

Start building realistic routes today

Routes that look fine on the surface often leak costs when it comes to execution. That’s because plans don’t account for the complexities of wholesale distribution.

The solution is to close the loop between planning and execution. By harnessing real performance data, you can build realistic routes with predictable costs.

Distributors and wholesalers don’t need to settle for routes that are just “good enough”. Descartes couples executional visibility with machine learning to build cost-effective routes that stand up to real-world constraints.

Get in touch with Descartes to start building profitable routes.