Medium- and heavy-duty fleets looking to adopt electric vehicles today are often stuck waiting years for charging depots to connect to the grid. This is due to constrained electricity grids and conservative utility planning practices.

When fleets and charging station providers looking to install charging sites submit a load request to their local utility, the utility planning team evaluates that request by looking at the worst-case highest feasible load demand already on the local distribution circuit and the worst-case load demand the new site could draw. If these combined loads exceed the safe operating limits of any local electricity grid component, the connection request is denied until infrastructure upgrades can be made. This denial happens even if the limit would be reached for just one hour a year at a time the fleet would never usually charge at.

Flexible service connections offer a faster and more efficient alternative. Flexible service connections allow sites to connect sooner in exchange for the customer limiting charging during constrained peak hours on the local distribution grid. With these peak hour load limits, the utility can ensure the local distribution grid remains within safe operating thresholds, enabling the deferral of infrastructure upgrades and cutting new site energization timelines from years to months at a time when upgrade queues can be lengthy.

Flexible service connections enable fleets to get their chargers energized faster and powering more electric vehicles on a site than would be possible under conventional planning, while allowing utilities to grow their rate base, increase their utilization of existing grid assets, and reallocate capital to higher-priority areas.

Exhibit ES1

RMI’s analysis in this report shows that most fleets can make the shifts needed by flexible service connections without impacting their operations. In addition, the majority of California utility Pacific Gas & Electric (PG&E) feeders could support an additional 10 heavy-duty trucks charging at connected sites through just four hours of load shifting by fleets in the late afternoon during a handful of peak summer days per year. Scaled across all PG&E feeders, this could theoretically support charging all the medium- and heavy-duty electric vehicles ever sold in California with no necessary grid upgrades. It could also absorb the more than 650 MW load growth expected from truck electrification in PG&E territory through 2035, as projected by RMI’s GridUp tool.

Currently, flexible service connections are available from a small number of utilities for a small number of fleets. However, more utilities are starting to pilot them, and the California Public Utilities Commission has recently mandated their expansion in the state. This report describes the real value available today for both fleets and utilities if policymakers, regulators, and utilities work together to enable the expansion of flexible service connections to more jurisdictions.

The United States is in an era of large electrical load growth. Demand for electricity has grown at a rate of 1.7% per year over the past five years, compared to just 0.1% per year over the prior decade. This demand is largely driven by data centers, but the electrification of major sectors of the economy is also a substantial part of this trend. All this new demand has resulted in lengthy queues to connect these loads to the grid as local electricity distribution systems become more constrained and require more upgrades. And delays are exacerbated by the fact that standard utility planning practices only allow upgrades once the demand has arrived as opposed to proactively as lines get congested.

Charging station providers and fleets looking to electrify are hit particularly hard by delays in site energization because charger installation and vehicle purchase timelines are often significantly shorter than the timelines utilities use to assess the impact of a new or upgraded connection onto the grid, such as for new construction. Although federal incentives have been rolled back, the combination of state-level incentives in places like California, the ongoing diesel and gasoline price shock, and the increasing availability of cost-competitive electric trucks means fleet electrification will accelerate, increasing the urgency for a solution to charger energization delays. Fortunately, electric vehicle charging — and especially the high-capacity charging required for electric heavy-duty trucks — provides an opportunity to shift conventional utility grid planning to a more flexible and cost-saving approach.

As RMI explained in our articles “Electrification 101: Enabling Truck Charging with Flexible Service Connections” and “Increasing the Number of Electric Trucks We Can Charge Through Flexible Service Connections,” when a new charging site wants to connect to the grid, conservative utility planning processes consider the highest possible load hour of the year on the local distribution grid and the highest possible load hour that site could conceivably reach. Then, if the maximum safe operating limit of grid equipment on the local distribution system — such as feeders — is reached at any time, the utility denies the request until costly upgrades can be completed. This happens even if that limit would only have been reached for just a single hour of the year at a time the site wasn’t even planning to charge vehicles.

By basing connection decisions on worst-case load scenarios, the current practice produces systemic and repeated inefficiencies that result in denying loads that could easily be managed through small operational shifts. Current rate-basing structures provide financial returns when utilities make new capital investments, such as upgrading feeders and substations in locations where grid constraints are increasingly likely. However, this approach is conservative in that it does not provide financial incentives for utilities to deliver electricity in the most efficient way with the highest utilization of all grid assets.

A new approach called flexible service connections (FSCs) is now being pursued by several utilities across the country, including PG&E with their FlexConnect program. FSCs enable faster grid connections and infrastructure upgrade deferrals, saving both utilities and customers money.

FSCs are formalized agreements between a customer and utility to treat the load limit of the grid more flexibly. The utility agrees to connect a project to the distribution grid in an area where constraints exist at certain peak load hours of the year, and the customer agrees to curtail their load during these peak periods as requested by the utility.

As exhibit 1 below shows, FSCs could enable a new charging depot in California to energize its site to its requested 4 MW of capacity most of the year, provided they curtail their load to 2.5 MW during peak daytime hours in June and July. In contrast, a traditional connection would have just given them 2.5 MW all of the time until an upgrade was completed. The FSC approach accelerates timelines for new site energizations for fleets, saves them money on site upgrade costs, and enables utilities to grow their rate base of customers with existing infrastructure, saving them money too.

Exhibit 1

While this solution is technically feasible for many sources of load connections on the grid, and is even applicable to energy generation sources, in this report we are focusing on the opportunity from electric trucks specifically. This is because electric trucks represent the perfect test case for FSCs. Fleets operate on tight profit margins, and so are responsive to fuel price signals when scheduling their operations, allowing for flexibility with their charging times. Past research by RMI has found that most fleets require less than half of their dwell time to charge at a 75 kW rate, and many fleets are now using chargers significantly more powerful than that. Additionally, fleet vehicles often plug in to charge with 40% battery life remaining, creating additional opportunities for flexibility.

In this report we aim to demonstrate the scale of the benefits FSC solutions can provide to both utilities and fleets, using real world feeder load data, truck telematics, and a truck charging optimization model developed by RMI to model the characteristics of where these solutions will work best and for what types of fleets. Rather than assessing specific program structures, the goal of this analysis is to explain FSCs' benefits more clearly to fleets and to demonstrate to utilities the scale of the potential market for this solution.

Flexible service connections are offered by very few utilities today due to the data and regulatory complexities in implementing them. Utilities require accurate and granular data on distribution system activity (such as via DERMS software) to move away from conservative planning restrictions toward flexibility agreements, something many utilities do not have today. The utilities that do have these capabilities have them at differing levels of granularity, and so different categories of FSCs are being piloted today.

The main feature of an FSC is the load limit enforced by the utility on the customer’s connection. Exhibit 2 below, from EDF’s 2025 report on program design considerations for FSCs, shows the full spectrum of load limits. Today, two of the major utilities offering FSCs are Pacific Gas & Electric (PG&E) and Southern California Edison (SCE).

PG&E’s FlexConnect program is on the right, more complex end of this spectrum, offering customers dynamic day-ahead limits on their load. This enables customers to get greater service for more of the year but creates operational uncertainty around when curtailment events may happen, which could create difficulties for a fleet that needs to charge its vehicles to complete a fixed schedule of deliveries.

Toward the left-hand side of the spectrum is SCE and its Load Control Management System (LCMS) pilot, which offers static seasonal and fixed-schedule load limits, for example giving customers full capacity for most of the year while curtailing capacity during all summer evenings, when loads tend to reach peak demand. This model may result in slightly less additional capacity being utilized than a dynamic one but is operationally easier for fleets to plan alternative charging schedules around.

Both approaches require customers to have an energy control management system that can respond to signals from the utility and ensure that load curtailment events or pre-agreed seasonal load limits are followed, giving the utility certainty that safe operating limits for the grid are not exceeded.

The boxes below show two case studies for how these competing approaches have played out with real fleets.

On the regulatory side, commissions need to oversee the development of a new kind of non-firm connection agreement for FSCs to scale. In particular, regulators will need to address several key issues:

• Tariffs and program rules: Utilities will require regulatory approval for new tariffs, clear eligibility criteria for which sites can enroll in FSCs, and transparent curtailment terms.

• Customer protections: Regulators will need to ensure that customers continue to receive safe and reliable service and are not disadvantaged by participating in an FSC program. For example, customers may need protection against losing their place in the regular interconnection queue if a long-term FSC ultimately proves unsuitable for their needs.

• Cost recovery and cost allocation: In most cases, local grid upgrades are partially paid for by the customer whose new load triggers them, while larger system upgrade costs are typically paid by the utility and recovered through rate-basing. For FSCs, key questions include who pays for an eventual upgrade if investment is deferred, and who bears the cost of the controls, communications, and software needed to implement the arrangement.

• Curtailment communication and enforcement: Regulators must determine how curtailment is communicated and enforced, as well as what performance data utilities must provide so customers can evaluate the risks and practicality of taking non-firm service.

• Duration and purpose of the arrangement: Regulators will also need to decide whether FSCs are implemented as a bridging solution until service upgrades are completed or as a more permanent approach to maximizing use of existing grid infrastructure. Each approach has different implications for cost allocation and potential customer reimbursement.

In February 2026, the California Public Utility Commission made strong movements toward addressing these issues by adopting a policy establishing a standardized Flexible Service Connection offering that PG&E and SCE are mandated to make available to customers facing distribution capacity constraints to accelerate energization timelines. This decision requires the utilities to expand their programs beyond pilots, establish a tariff, and collect performance data to aid in refining and evaluating the cost-effectiveness of the programs. The regulation also states that any customer who enrolls in these programs will not lose their place in the load queue, reinforcing FSCs as a bridge solution in the near term, while leaving the possibility open for long-term deferral of distribution upgrades.

Quantifying the Benefits

When Are the Benefits Greatest?

Where Are the Benefits Greatest?

How the Benefits of Flexible Service Connections Vary by Utility

Increasing the flexibility limit from 1 MW to 3 MW only resulted in marginal gains in our analysis. This is due to the median peak prominence in PG&E being below 2 MW as discussed earlier. Greater flexibility limits may be necessary in real-world situations where utilities need to be more conservative to prevent blackouts, but these greater curtailment levels make it more difficult for fleets to comply with FSC requirements through managed charging alone.

By adding DERs to their depot, as PepsiCo and Terrawatt did in the case studies, fleets can comply with load-shifting requirements with relative ease and retain additional operational flexibility to charge when they want. This is particularly valuable for fleets under dynamic FSC agreements like those in PG&E. Adding on-site battery storage and solar can add significant up-front capital costs to a project, but past RMI analysis has found this to be a profitable strategy even without FSCs in jurisdictions where net metering or feed-in tariffs exist.

This analysis has shown that real benefits to the grid and to fleets are possible if utilities and regulators come together to expand access to FSCs. And benefits are strongest when utilities and regulators work with fleets to design programs that work for them and their operations, with clearly explained, fair, and predictable curtailment rules that incentivize fleet enrollment.

More state utility regulators need to follow California’s lead in supporting the development of standardized offerings for FSCs that include incentives for utilities to scale FSCs. In January, the CHARGED initiative released an implementation guidebook to support utilities and regulators in this journey. Implementing these changes enables the grid to make the most of existing assets and puts downward pressure on electricity rates, supporting energy affordability.

For fleets, FSCs represent a huge opportunity to electrify faster, but they do need to ensure their operations are suitable for the constraints inherent in FSCs. Predictable seasonal limits such as those piloted by SCE are easier for fleets to plan their operational schedules around and should be the approach utilities start with when expanding these programs. There is a significant opportunity to educate fleets on how to manage operations in a way that enables flexible charging. Even small amounts of charging flexibility, shifting charging by just a couple hours, can often unlock large gains for truck electrification.

Utilities should proactively identify the feeders with the maximum possible benefits from FSCs, such as those near population centers and those nearing their capacity limits, and engage fleets located near those feeders that are considering electrifying on participating in FSC programs. Additionally, as more data is gathered and confidence grows in FSCs, utilities should move to turn them from a bridge solution for deferring upgrades to a tool for permanently avoiding unnecessary upgrades.

RMI developed an optimization model to find the additional number of trucks that could be electrified using flexible service connections. Flexible charging is allowed during defined hours within a day to assess which hours provide the most value for trucking electrification.

Public feeder data from Pacific Gas & Electric and National Grid was used as the inputs to show geographic differences in grid behavior. Feeder load scenarios for the 90^th percentile of load demand month-hours was used to represent the conservative utility planning approaches.

Appendix Figure 1

Truck load curves were built using Geotab telematics data as a baseline and then assuming either unmanaged charging, slow charging, or overnight charging, as shown in the figure above. Unmanaged charging assumes trucks begin charging at full speed as soon as they return to their depot. Slow charging spreads power use evenly over vehicles’ dwell time, and overnight charging delays 80% of charging to the 9 p.m.–7 a.m. window. Distinct load curves were used for urban and regional-haul fleets to show the difference in energy needs and power demands between the two, with urban trucks driving just 100 miles per day on average and regional trucks 300 miles.

Appendix Figure 2

regional heavy-duty truck hourly load curves illustration of how our flexibility model works for a six

To assess the benefit of flexible service connections, a business-as-usual (BAU) approach assessed the number of trucks that could be accommodated on each feeder, using the standard truck load curve, without exceeding the feeder headroom. The optimization model then allows for truck charging load to be shifted within a set number of hours, and up to a maximum power level to calculate how many additional trucks could be charged on a given feeder above the BAU value. For instance, to avoid straining a feeder, a truck may charge more in the hours immediately before and after the time when the feeder’s available power capacity drops. By evaluating which hours of the day allow for the most additional electric trucks on each feeder, behaviors across the larger utility service territory can be identified.

Analysis from a past RMI study on fleet time stopped at domicile underpinned the core logic of our model.

Appendix Figure 3