Data-Driven Utilization Forecasting: The Overlooked Pillar of EV Charging Success

(Part 2)

From Static Estimates to Scenario-Based Forecasting

Traditional demand estimates give you a snapshot. But successful EV infrastructure needs a dynamic picture — one that accounts for real-world variation in user behavior, daily operations, weather and local context. This is where data-driven utilization forecasting becomes indispensable.

In part 1 of this 2-part series, we explored why utilization forecasting is a foundational component of EV charging project success and how common industry practices often fall short. We examined the costly consequences of misaligned infrastructure, the limitations of high-level demand models, and how macro trends like fleet electrification and capital tightening demand more granular, site-specific planning.

Now, in Part 2, we turn to solutions, focusing specifically on how advanced utilization forecasting works, what kind of inputs make it effective, and how system-oriented platforms (like Brightmerge) supporting such forecasting equip EPCs and CPOs to plan smarter, build leaner, and operate with confidence.

Key Forecasting Inputs That Matter

So what makes data-driven utilization forecasting so different?

Stated simply, instead of relying on fixed assumptions (e.g., "each charger will be used 3.5 times per day"), a data-driven utilization forecasting engine would build multi-variable, site-specific scenarios that simulate how charging assets will perform over time, including at different hours of the day, on weekends vs. weekdays, or across seasons. This shift in approach transforms planning from speculative guesswork to evidence-based decision-making.

Ideally, utilization analytics are powered by a deep array of data inputs that reflect the true complexity of how EVs interact with infrastructure. These inputs include:

1. Driver and Fleet Behavior

• Dwell times: How long do vehicles typically remain parked while charging?

• Turnover rates: How frequently do new sessions begin?

• Fleet shift schedules: Are vehicles returning at predictable intervals? Is charging spread out or concentrated?

• Charging windows: Are vehicles available for charging overnight, during lunch breaks, or in tight operational windows?

• Peak versus off-peak charging: What are the implications [financial, grid stability, etc.] of instantaneous demand calculations regarding when to charge?

By understanding these variables, it is possible to forecast session distribution across time and ensure charger availability aligns with actual usage patterns.

2. Local Traffic and Trip Patterns

• Origin–destination data and traffic flow influence how often vehicles stop at charging stations.

• For public charging, proximity to high-turnover zones like shopping centers, airports, or logistics corridors is key.

• For fleets, depot access routes and delivery paths shape charger placement and power requirements.

These insights help determine not only where to place chargers, but also what kind - and how many - are needed.

3. Peak Demand and Load Shaping

• What are the most intense periods of usage at a given site?

• Can charging be staggered to reduce strain on the grid or minimize demand charges?

• Are there patterns of queueing or underuse at certain times?

• Are there microgid solutions that make economic sense (e.g.: battery storage during peak electricity demand conditions)?

Forecasting session peaks and pairing that with grid-aware charging strategies enables optimal power provisioning, better utility engagement, and smart load management.

4. Growth and Electrification Trajectories

• How quickly will EV adoption scale in a given region?

• Are fleets expanding, contracting, or transitioning vehicle types?

• What new user groups (e.g., rideshare, commercial delivery, public) are expected?

Factoring in regional EV growth curves, local policy mandates, and even adjacent development activity to simulate future site throughput, ensures infrastructure is neither overbuilt nor quickly obsolete.

Real-World Application: A Fleet Charging Example

Imagine a mid-sized fleet operator planning a new depot for 60 last-mile delivery vans. Without precise utilization forecasting, planners might install 60 chargers, meaning one for each vehicle. But that’s both rarely necessary, and rarely efficient.

With a system-focused approach to analytics such as that adopted by Brightmerge, the team can model:

• Vehicle return times (e.g., 80% of vans return between 6–8 p.m.)

• Dwell durations and required state of charge by morning

• Staggered charging potential using Level 2 and DCFC combinations

• Grid constraints and possible load-shifting to off-peak hours

• Economics for battery storage

The result? Instead of 60 chargers, the depot might require only 25 strategically scheduled chargers, possibly saving hundreds of thousands of dollars in hardware, trenching, and interconnection costs. And while the numbers used in the example above are representative of real-world cases, we changed the numbers to protect client identity even as real-world analyses for clients have proved these savings, saving them significant amounts by uncovering how smart scheduling of their fleet reduces charger requirements and energy costs. 

But the real value - beyond the savings - can be found in the fact that the site is designed for reliability, scalability, and ROI from the outset.

Beyond ROI: Operational Efficiency and User Experience

Utilization forecasting doesn’t just protect capital - it also shapes the day-to-day performance of EV charging infrastructure.

• For operations managers, it enables better scheduling, maintenance planning, and uptime assurance;

• For fleet supervisors, it ensures vehicles are ready to go when needed, without excess wait times or queueing;

• For site designers, it informs power sizing, charger mix (DCFC vs. Level 2), and even physical layout.

Ultimately, this leads to smoother workflows, fewer surprises, and a better experience for both drivers and administrators.

How Brightmerge Integrates Intelligent Forecasting

Brightmerge’s platform offers utilization forecasting as a core building block of its broader site modeling solution, delivering:

1. Financial Modeling Integration - Utilization forecasts feed directly into revenue projections, TCO analysis, and ROI timelines — giving finance teams confidence that sites will meet investment thresholds.

2. Grid-Aware Planning - Forecasted demand helps right-size electrical infrastructure, reduce overbuild risk, and avoid costly delays or redesigns due to underestimated capacity needs.

3. Fleet Charging Scheduling - Smart scheduling of vehicles to take advantage of their individual route requirements, battery storage capacity, TOU tariffs, etc can reduce the number of chargers required, the grid support needed, and the ongoing amount of energy consumed and ensuing cost of energy, amongst other parameters

4. Iterative Design Support - Forecasting is not a one-time activity. Brightmerge allows for dynamic re-forecasting as project parameters evolve — such as when fleet profiles shift or local traffic data updates.

5. Multi-Stakeholder Alignment - The platform presents clear visualizations and scenario comparisons that bridge conversations between planners, utilities, fleet managers, and investors.

   
   

Utilization Forecasting as a Strategic Lever

Utilization forecasting may seem like a narrow, technical topic, but as discussed in part 1 and above, it influences everything from:

• How many chargers to install

• Where to site them

• How to size interconnection

• What infrastructure to prioritize

• How to ensure long-term profitability and performance

It is, in many ways, the glue that holds together effective EV infrastructure strategy.

And critically, it doesn’t operate in isolation. It connects to the other 4 pillars outlined in the 5-Pillar Risk Mitigation Framework presented in the first article in this series, namely:

• Upfront Financial Modeling

• Grid-Aware Smart Charging

• Agile Infrastructure Planning

• Financialization of Sustainability

Together, these pillars create a systems-level approach — one that moves EV charging projects beyond trial-and-error toward scalable, data-driven excellence.

Conclusion

As we move into the next phase of EV infrastructure build-out, the margin for error narrows. Projects must deliver not just capacity, but performance. Not just hardware, but ROI.

Utilization forecasting is one of the most powerful - and underutilized - tools in the planner’s toolbox. With the right data, analyzed from a systems-level perspective, EPCs and CPOs can design charging sites that are leaner, smarter, and built for long-term success.

____________________________________

To better understand how Brightmerge leverages data-driven utilization forecasting, schedule a call with us.