According to the North American Council for Freight Efficiency (NACFE), the Total Cost of Ownership (TCO) components of a Class 8 Battery Electric Vehicle (BEV) are, from highest to lowest: Labor, Vehicle cost, Fuel, Taxes and fees, maintenance and repair, insurance, payload, and financing.
This makes sense. But many TCO calculators and consultants don't account for the labor component accurately, as it is signifcantly related to delivery network design and vehicle routing. In many instances we see that they use averages or historical telematics data that are deeply flawed as inputs to what the real customer-driven fleet requirements are throughout the course of a year. For example, if you have to cover 10,000mi of deliveres from a DC and that would require 10 x 1,000mi rig routes, swapping in 5 BEVs with only a 500 mi range means you need more break and charging time, and potentially more driver shifts. So 10 full time driver shifts in diesel trucks could now be 12 or even 15 driver shifts in a mix of ICEs and BEVs.
Then the questions become:
- What is your optimal mix of ICE and ZEVs for lowest TCO today?
- What is the optimal mix in a year, in 5 years, and in 10 years that achieves minimum TCO and achieves your Net Zero goals?
- What Capex and Opex budgets are required to achieve it?
In addition to vehicle and infrastructure Capex and Opex, it's critical to look at the reduced range of the vehicles and their impact on routing and labor costs dynamically throughout various time periods. This can be done by combining dynamic route optimization and network planning with TCO inputs and performing higher level optimizations to reach suggestions for where and when ZEVs can be substituted into the network.
In this article, we'll be ignoring some of the other TCO components which are well covered in articles like this one from 7Gen and this great Youtube video. And to learn more about the opportunities and challenges of Class 8 ZEV transitions, check out this article by Logisyn Advisors.
Here we will show you how one customer took the total customer demand of the delivery network from one distribution center across an entire year and calculated the optimal delivery route requirements using real traffic and other inputs, and then simulated that network with different ZEV mixes to determine the optimal fleet makeup at the lowest TCO and implementation headache.
To acheve this, Adiona's rapid ML-powered delivery routing platform performed hundreds of simulations quickly and then processed higher level optimization algorithms to choose the ideal (lowest TCO) scenarios. This was done over short (1 year) and long (10 year) horizons so the organization could "begin with the end in mind" and plan their POC phase with clear goals and KPIs.
This large retail customer is transitioning from Class 8 diesel vehicles to a mix of Fuel Cell Vehicles (FCV) and Battery Electric Vehicles (BEV) to achieve their Net Zero goals. FCVs are more expensive up front and the refueling infrastructure is more expensive, but they achieve significantly longer range than BEVs. So we start by looking for the 'low hanging fruit' of which delivery routes can accomodate each vehicle type with today's operational requirements (customer demands, time windows, replenishment templates, order patterns, seasonality, etc) and how that might change over time.
Adiona's tool requires at a minimum: The up front vehicle costs for each type, which may include the purchase price and any technology-specific costs such as driver top up training. It also takes in the per mile cost of each vehicle, the route distances, the route schedules calculated over at least a historical year, and the projection length in months or years. It's also easy to add in other costs for infrastructure as flat costs or included in the per-mile cost, such as fuel rates.
The TCO Optimization tool will then calculate the optimal mix of vehicle types which may include the target vehicles and also legacy vehicles if the constraints for using a newer target vehicle can't be met at the optimal cost. The output will look like this.
(The outputs will include other parameters and can be run at scale to cover many different scenarios.)
With these scenarios in hand, and the costs calculated for fuel and maintenance from traditional analyses, a few different visualizations were created.
Here is a comparison of three scenarios of TCO per mile and the relevant vehicle type mix that ensures all of the delivery routing requirements are achievable in a territory with a single DC and over 33,000 miles of delivery routes to cover from this customer's DC.
The BEVs have a maximum range of <500mi between charges, so assuming we don't have a charge infrastructure on road yet (because we really don't for Class 8 trucks), we let the routing tell us how many vehicles with that range could possibly be used and still satisfy the operational needs. Likewise we look at the FCVs with 800mi range, and the remainder of longer hauls will still have to be ICE. We see that depending on the cash outlay for ZEVs up front, we can optimize the selection of different vehicle mixes and still achieve lower operating costs per mile.
Scenario 1: Lowest TCO, Highest Capex
For a 1 year horizon, the optimal mix for lowest TCO is 40 BEVs, 16 FCVs, and 7 ICE (diesel) vehicles.
Scenario 2: Higher TCO, less up-front Capex
If we want lower Capex (and lower operational headache), we can acheive next-best TCO with 39 BEVs, 17 FCVs, and 9 ICEs.
Scenario 3: Highest TCO, easiest implementation
Our lowest hanging fruit scenario is 37 BEVs, 18 FCVs, and 11 ICEs.
As you can see, the benefit of this approach is that it provides clear, evidence based results and actions that will work in a real network. Unlike Excel-only analyses, optimization-based approaches like this allow for faster Net Zero transition by getting new wheels on the road at the right times to suit the real operations.
Get in touch with us to discuss how to use our tools for your own TCO analysis.