Route Optimisation

Use AI to optimise your delivery network and reduce carbon emissions

Reduce delivery times and costs
Reduce carbon emissions
Improve vehicle utilisation

Route optimisation

Table of contents

Most route optimisation tools have limited capabilities—planners are not able to flex constraints and configure what should be prioritised in the schedule optimisation. This limits the ability to test out different hypotheses for how to improve network efficiency (for example finding the best day of the week to do the job on) and restricts organisations in meeting their specific KPIs (for example optimising the network to achieve a balance of reducing vehicle costs and carbon emissions).

Traditional route optimisation takes weeks to create strategic schedules which is laborious for the team, slows time-to-value and is prohibitive in generating different scenarios to assess. Regular operational changes mean that routes are inefficient until another schedule can be created. For daily and dynamic schedules, there is not enough time to plan routes optimally and incorporate late changes to requirements.

When route optimisation tools produce suboptimal results, planners make schedule adjustments which introduces individual bias, subjective debate and a lack of visibility around what changes are being made and why. Drivers and crew can decide to make changes to their route whilst on the road which further dilutes their efficiency and reduces transparency of operations.

AI for route optimisation

Route optimisation is the process of planning the most efficient and effective routes for a fleet of vehicles to travel to service deliveries, customer appointments, equipment maintenance appointments or other types of jobs. It involves taking into account a variety of factors, such as the location of stops, vehicle size and capacity, crew skills, traffic conditions and time constraints.

It's a classic problem in optimisation—and is the generalised version of a problem referred to as the Travelling Salesperson Problem.

Route optimisation is a complex task that only gets more complex as you deal with larger fleets, more constraints or increasingly complicated schedules. Hence, traditional methods of route optimisation can be time-consuming and—as complexity increases—inaccurate, as they may not be able to take into account all of the relevant factors.

AI route optimisation solutions can help businesses to overcome the challenges of traditional route optimisation methods. These solutions use AI to analyse large datasets, including job data, traffic data and vehicle data. This information is then used to generate optimal routes for large, complex networks.


AI route optimisation solutions provide a number of benefits, including:

  • Reduced costs: By reducing the number of vehicles required, the number of staff shifts and the amount of fuel used.
  • Reduced CO2 emissions: By reducing the number of miles driven and potentially type of vehicles used (e.g. moving to electric vehicles).
  • Improved asset utilisation: By enabling a smaller number of vehicles and staff to do more jobs, deliver higher loads and do more miles, reducing overall asset requirements.
  • More product delivered: By enabling higher fill rate of vehicles and reducing the risk of unbalanced routes where there is more product assigned than capacity on the day.
  • Increased customer happiness: Fewer delays, through more realistic schedules that better take into account customer SLAs and route time requirements.
  • Meet regulatory compliance: Through robust and feasible schedules that comply with constraints which include regulatory requirements (e.g. driver working time directive which states staff cannot work for more than a certain amount of time without breaks).‍
  • Better managed operational risks: Through scenario modelling of risks and disruptions to put in place contingency planning.

Who it’s for

AI route optimisation solutions can benefit businesses in a wide range of industries, including retail, logistics, transportation and field services. They benefit businesses of all sizes, from small businesses to large enterprises. Though, the more complex the delivery network, the higher the ROI an AI solution is likely to produce.

Some of the specific roles that can benefit from AI-powered route optimisation solutions include:

  • Fleet managers
  • Dispatchers
  • Scheduling managers
  • Transport planners
  • Transportation managers
  • Logistics managers
  • Delivery managers

What it can be used for

AI route optimisation solutions can be applied throughout the planning and routing process, from strategic design and modelling that is done on an infrequent basis at an entire network or depot level to daily and near real time routing of vehicles on the road.

  • Network design: Identifying optimal location of depots or service hubs, fleet size and mix, staffing requirements and mix.
  • What-if modelling: Scenario planning for different demand patterns, job location changes, constraint flexing such as changes to customer SLAs or job requirements and contingency planning such as industrial action.
  • Resource and commercial planning: Defining vehicle and shift requirements, assessing costs and revenue, evaluating customer pricing, forecasting ESG impacts.
  • Strategic / framework planning: Creating typical schedules and route allocation to inform day to day operations.
  • Daily routing: Generating schedules and routes per staff and vehicle specific to daily requirements for immediate execution.
  • Dynamic: Adapting routes that are being driven in near real time to take into account job cancellations, emergency jobs, road closures or diversions.

How it works

A typical Datasparq route optimisation solution works by analysing the following types of data:

  • Historical job data (such as delivery or appointment times, distances, and traffic conditions)
  • Vehicle data (such as vehicle size, capacity and range)
  • Job data (such as the location of stops, arrival and departure times, and the staff skills requirements)

This data is then used to train a machine learning model which is configured to adhere to that business’s specific constraints. The model can then be used to create optimal routes for individual vehicles, or for entire fleets of vehicles.

Getting started

As outlined above, by optimising delivery routes, AI solutions can help businesses to reduce costs, improve customer service, increase productivity and reduce their environmental impact.

Contact us today if you're interested in learning more.

No items found.
No items found.

Let's find a time to talk

Contact us

Ten ways to use AI in supply chain & logistics

Datasparq make AI a reality for Fortune 1000 clients such as GXO.

Download this guide to...

✓ Learn 10 key use cases of AI in supply chain & logistics
✓ See real life examples of value-driving AI & data science projects
✓ Find out exactly what you need to get started

⚠️ Please enter a valid business email address
No spam. Ever.
Thanks for downloading the guide
Oops! Something went wrong while submitting the form.

When you're ready, we're here

Contact us