Blog post

Effective EV Route Planning: How to Overcome Key Industry Challenges

How emerging tech solutions can bridge the gap between physical constraints and new operational demands

Updated: August 21, 2023 10 mins read Published: May 02, 2023

Fleet electrification is accelerating. Rising fuel prices, nearing net-zero targets, and the never-ending quest for cost optimization are getting us closer to green mobility.

The EU parliament has voted to ban new sales of carbon-emitting petrol and diesel vehicles by 2035. The UK government has an even more ambitious target of prohibiting sales of all new petrol and diesel cars by 2030. By the same year, the global electric vehicle market is forecasted to reach $2.7 trillion, up from a “shy” $543.8 billion in 2022.

Regulatory changes won’t just affecting regular drivers. The implications are much more substantial for commercial fleets, both private and state-owned.

Decarbonization of the global transportation sector will be a bumpy road. Ambitious pilot programs have been stalled by practical obstacles such as insufficient EV charging infrastructure, limited battery capacities, and a wider range of operational issues.

Three key challenges of EV route planning and navigation

EV fleets come with a host of unique needs: energy procurement, charge scheduling, battery servicing, and context-aware route planning.

To accommodate, companies will need to create new fleet management flows.

Effective EV Route Planning: How to Overcome Key Industry Challenges

Source: AMPLY — Managed Charging Accelerates Cost & Health Benefits of EVs

Yet, when it comes to action, mobility leaders often focus on physical limitations rather than the slightly less evident digital solutions that can be implemented with the right technology partner.

Learn how fleet operators can accelerate the industry transition through developing EV charging infrastructure

Read more

Energy procurement for EVs

In 2021, global EV fleets only consumed about 50 TWh of electricity (less than 0.5% of the total final electricity consumption worldwide). However, energy consumption will grow in proportion to fleet sizes.

By 2030, EV fleets in the EU alone will demand 187 TWh. That’s over half of the energy the region consumed in 2022.

In other words, EU countries will need to increase energy production and procurement by at least 50% to cover the demand of e-mobility and other industries. Given the current constraints in infrastructure and energy sourcing strategies, this will likely be challenging.

The European Commission already introduced a REPowerEU plan, which would accelerate the region’s share of renewable electricity sources. However, the program allocates an insufficient share of renewable energy for the transportation sector.

In light of fleet electrification, fleet managers are naturally concerned about energy procurement along with rising energy costs, which can substantially increase EV fleet operating costs and constrain fleet expansion.

The better news? EV fleet operators can prepare for a greener tomorrow today.

You can start measuring energy consumption by different EV models to benchmark present-day consumption and evaluate future demands. EV fleet management systems can be upgraded with a predictive energy consumption estimator (powered by big data and machine learning) that can dynamically calculate current consumption rates for each EV based on model, driving style, and terrain.

This way, you can estimate energy costs for each route. You can then use these insights to model various operational scenarios. For example, you can:

  • Evaluate the feasibility of servicing new long-haul routes
  • Estimate commercial viability of different destinations
  • Determine the best-performing EV models for different terrains
  • Model future energy procurement needs under different fleet expansion scenarios

With precise data at hand, your organization can then liaise with energy sector players and governments on various sourcing scenarios. For example, you can negotiate deals where you reserve a fixed number of MWh/GWh per month for a discounted rate.

EV fleet owners can also become new energy market players by commercializing their fleets’ energy generation and storage capabilities.

To meet future energy needs, larger mobility companies could invest in off-the-grid power generation (e.g. on-site solar installations). Tesla, for example, continues to expand its network of Supercharger stations powered exclusively by solar panels.

Alternatively, companies could procure renewable energy directly from off-grid generation facilities, which could offer companies up to $8.6 billion in cost savings due to differences between retail and wholesale energy prices.

Batteries with larger capacities could soon allow fleet managers to purchase power during off-peak hours. They could then use the accumulated energy to recharge EV fleets during peak loads.

Finally, fleet operators could double as energy traders on the side with the help of vehicle-to-grid (V2G) connectivity — a system that allows EVs to sell stored energy back to the public grid whenever there’s high demand. Fleets with predictable charging patterns and energy usage patterns (such as public electric buses) could sell extra energy during off-duty hours to generate extra profit for owners.

Efficient EV charging

It’s possible to get enough energy at competitive prices. But the e-mobility industry also needs sufficient EV charging infrastructure at parking locations and en route to make the transition worth it. And more importantly, fleet managers need to figure out effective charging schedules.

An EV with a 60kWh battery takes under 8 hours to charge from empty to full with a 7kW charging point. Level 3 DC fast charging stations can add up to 250 kilometers (~155 miles) of range per hour. Some vehicles can even achieve 80% in 30 minutes with a Level 3 charger.

These charging times are for light-duty and passenger vehicles, however, whereas most fleets have medium- and heavy-duty vehicles.

Volvo argues that a good charging strategy can compensate for the current limitations of AC (on-board) and DC (external) chargers:

If you’re in regional transportation, where you have long distances to cover, then you may need to find charging opportunities along your route. For example, if a truck regularly commutes between two set depots within 300 km, a charging station at each should be enough to make 24/7 electric vehicle operation possible.

DHL has successfully tested such a strategy with electric Volvo FE and Volvo FL trucks in Sweden and now plans to expand its e-truck fleet.

Still, for larger fleets, especially public transport fleets, managing EV charging will require meticulous coordination and planning. Charging schedules must correspond with workers’ shifts and pre-planned delivery schedules. Moreover, simultaneous charging must not overload local grids, causing disruptions and energy cost spikes.

Investing in a smart EV route planner is the first solution.

This way, you can plan routes that include en-route charging for long-haul journeys using available Level 1/2/3 stations. In the future, EV planners could also account for overhead catenary systems and wireless induction systems that are currently being developed.

At parking locations, EV charging management software can help design the optimal cadence for charging vehicles by modeling charging schedules based on planned travel schedules.

EV charging management software also aggregates historical and real-time charging data, which can provide ample insights for each charge point. This way, you can remotely grant access to the correct vehicles, monitor the charging process, and get notified about any servicing issues.

Likewise, you can program how much electricity is allocated towards each charged vehicle (based on dispatch management plans) to engage in time-of-use arbitrage. Plus, you can remotely enforce rules against overcharging or unauthorized (out-of-queue) usage.

Learn how to get operationally ready for fleet electrification

Read more

EV battery servicing

EVs don’t need as much physical maintenance as internal combustion engine (ICE) vehicles, requiring no oil or engine checks. Already, the annual tax and maintenance costs for EVs can be up to 49% lower than for equivalent ICE models, while refueling can be up to 58% cheaper.

Yet servicing (and replacing) EV batteries remains complex and expensive. The average lifespan of a light-duty EV battery is 8 years (or 160,000 kilometers / 100,000 miles), which isn’t quite sufficient.

To compete with conventional vehicles, electric-drive vehicles, and their batteries must perform reliably for 10 to 15 years in a variety of climates and duty cycles.

National Renewable Energy Laboratory

Moreover, the challenge of safe EV battery recycling remains unresolved. OEMs will need to create a process for collecting and safely disassembling batteries that contain toxic lithium-ion compounds. Given battery production costs, it would also make both economic and environmental sense to implement measures for battery refurbishing or repair.

Newer technologies such as fuel cells will substantially extend the lifespan (and capacities) of electric vehicles in the future. At present, however, fleet operators can maximize the longevity of their fleets with EV battery analytics.

By creating digital twins — data-driven EV battery models — fleet operators can predict the battery lifespan of in-service vehicles based on their model, age, and usage patterns. With this data, fleet operators can optimize vehicle charging patterns to extend battery life and better coordinate battery servicing for e-fleets.

Advanced battery control algorithms, implemented at the hardware level, can further facilitate battery performance. Analysis at the cell and pack levels can provide extra insights into the reasons for battery degradation in order to mitigate the effects with real-time electrochemical controllers. Promising research in this area is already underway.

Building a context-aware EV route planner app

The above challenges affect EV performance. When successfully resolved, the only issue that remains is EV route planning.

Compared to petrol cars, there are more variables in determining the feasibility (and costs) of completing certain routes:

  • EV model and range
  • Weather conditions
  • Terrain type
  • Traffic conditions
  • Vehicle load
  • Battery load
  • Driving style

Research suggests that each of the above factors affect EV energy consumption (and therefore range) to a different extent. For example, driving an EV on a flat route with exposure to 8.5 km/h winds has been found to increase energy consumption by 14%, while driving on a longer hilly route with a lower wind speed increased consumption by only 5%. With a wind speed of 50 km/h, energy consumption increased by 31% and 15% (respectively) on the same routes.

The impact of wind in different terrain complicates the well-known Traveling Salesman Problem in route planning — aka finding the shortest, most cost-effective way to travel among several destinations.

The electric vehicle routing problem (EVRP) has been solved

New and better EV routing solutions emerge regularly thanks to progress in machine learning (ML), deep learning (DL), and reinforcement learning (RL).

Our mobility team has also developed an EV route planning solution based on machine learning algorithms and sensor-based analytics. Here’s how it works.

Decision tree algorithm

A decision tree is a type of supervised machine learning method (i.e. one where you provide initial input) that uses classification and regression for decision-making.

Effective EV Route Planning: How to Overcome Key Industry Challenges

Decision tree algorithms are well-suited for tasks like EV route planning because they can help you rapidly evaluate multiple go/no-go scenarios based on predefined conditions such as the EV model, current charge rate, and daily weather forecast.

Based on the input data, you can design a list of critical paths — conditions that must be met within the project (a planned EV trip) that are critical to the project’s completion (getting from point A to point B).

Learn where to look to profit from fleet electrification

Read more

Predictive critical path calculations

An ML-powered route planner for electric cars relies on multiple data points to estimate the feasibility and effectiveness of any selected route.

These include:

  • Standard navigation parameters (point-to-point distance, total distance, etc.)
  • Operational parameters (deadheading, conflicts with driver shift hours, etc.)
  • Real-world conditions (wind speed, traffic congestion, etc.)

Using all these inputs, the algorithm designs a set of routes that fit the selected planning criteria. Afterwards, dispatch managers (and drivers) can approve, adjust, or reject the suggested options.

For example, an integrated EV trip planner our team recently created for an OEM automatically calculates the best route based on any number of stops and taking into account required charging. To avoid unnecessary detours, charging points closest to the driver’s destination are automatically added to the route. Moreover, an estimated time of arrival (ETA) and estimated charging time (ECT) are always visible on the dashboard. Both are updated dynamically based on driving patterns (such as when entering a curvy road or getting extra wind exposure). The electric car trip planner can also start heating the vehicle battery as the arrival to the charging station approaches to achieve better charge quality.

Sensor data fusion

The most advanced electric vehicle route planners also factor in fleet telematics data.

EVs are equipped with multiple sensors that allow you to get a first-hand look at real-world vehicle performance from afar. Helpful data points include thermal runaway detection, liquid cooling, and battery charge status.

Telematics data provides insights into the vehicle’s actual performance under specific weather conditions and given a particular terrain, load, and driving style.

Sensor data fusion is the process of combining information from individual sensors into a consolidated stream of insights. With the latest EVs, it’s possible to perform sensor data fusion on the edge (in the car) and immediately generate real-time instructions for drivers. For example, you can warn the driver about excessive acceleration or suggest they make a charging stop at the next station because of the current weather conditions.

Intellias recently helped deploy an ADAS system, which relies on location-based services to automatically optimize battery power consumption and regeneration. Based on the ADASIS v3 protocol, the system enables ADAS apps to access data on the map route, vehicle position, and vehicle speed for smarter emergency braking, better predictive headlights, and reduced power consumption.

Rooting for your success with EV route planning

The transition to electric fleets will be anything but simple. However, EV route planning is another tool at your disposal. EVs have ample (and readily accessible) telematics data that fleet operators can use to devise effective one-off and recurring routes, schedule dispatches, and manage e-fleet charging.

Present-day progress in sensor data fusion, machine learning, and deep learning also allows you to take advantage of data to optimize e-fleet battery performance, charging patterns, and, ultimately, the commercial profitability of transportation services.


Contact Intellias to learn more about our expertise in delivering navigation, mapping, and operational software for global innovators in the mobility sector.

Rate this article
5.0/5.0 Thank you for your vote. 59024 0bb006b418
How can we help you?

Get in touch with us. We'd love to hear from you.

We use cookies to bring best personalized experience for you.
By clicking “Accept” below, you agree to our use of cookies as described in the Cookie Policy

Thank you for your message.
We will get back to you shortly.