What problem does it solve?
Vegetation growing into or falling onto lines interrupts supply. The NERC transmission vegetation standard notes that major outages and operational problems have resulted from overgrown vegetation interfering with transmission lines, and AiDASH calls vegetation one of the grid's biggest threats. AiDASH says vegetation programs have for decades largely followed a fixed formula: trim a set share of the system each year and repeat the cycle. Its case studies describe Entergy on a standardized five year cycle, maintaining about 20% of its system each year, and National Grid's Massachusetts network on a typical five year cycle, where deciding whether a circuit needed pruning took manual field reviews.
AiDASH notes that growth rates vary by region, species, weather and circuit, so a uniform cycle means unnecessary work in some areas and elevated risk in others. According to AiDASH, National Grid had deferred work for four years running rather than fund it, and in the year after pruning its average circuit saw only an 8% reduction in customers interrupted and no significant improvement in tree events or customer minutes interrupted. The same case study describes rising costs for routine maintenance and strict regulations to prevent outages and manage fire risk.
- NERC's transmission vegetation standard FAC-003-5 states that major outages and operational problems have resulted from interference between overgrown vegetation and transmission lines.FAC-003-5 Transmission Vegetation Management (2021)
How does it work?
- Image the whole network. Satellite imagery, supplemented by aerial or lidar data where needed, covers every span, repeated as often as the budget allows.
- Measure the vegetation. Computer vision models identify trees, their height, their distance to the conductors and signs of poor health, span by span.
- Predict growth and risk. Models combine species, growth rates, weather and outage history to estimate when each span will become a risk.
- Plan the work. Circuits are scheduled for trimming when their risk warrants it, hazard trees are prioritised for removal, and budgets are allocated where they avoid the most outages.
- Dispatch and audit. Work goes to contractors with maps, and new imagery checks that the work was done and done well.
- Measure reliability. Tree related outages, customers interrupted and minutes interrupted are tracked per circuit, and the results tune the risk model.
- Audience
- Employee facing
- Autonomy
- Copilot
- Adoption
- Early adopters
- Channels
- Internal tools, Mobile app
What is it worth?
Benchmarks are computed from the public deployments below: one data point per organization per KPI, with who made each claim.
| KPI | Median | Reported range | Data points | Claimed by |
|---|---|---|---|---|
| Cost savings | Not pooled | USD 2 million | 1 | 1 vendor |
Value drivers: Lower cost to serve, Risk and loss reduction, Customer experience, Employee productivity.
Indicative value
An electric distribution utility with 20,000 line miles
USD 600,000 to USD 4 million
Vegetation management spend avoided or redeployed per year
How this is calculated
Formula: vegBudget * savingShare. The low scenario uses every low input, the high scenario every high input.
| Input | Low | High | Basis |
|---|---|---|---|
| Annual vegetation management spend vegBudget, USD per year | 20,000,000 | 40,000,000 | Editorial assumption for a network of this size, replace with your own budget. |
| Share of spend saved by trimming on risk instead of on a fixed cycle savingShare, fraction of vegetation spend | 0.03 | 0.1 | Editorial assumption, deliberately below the 20% expense reduction AiDASH claims in its own marketing on the Entergy page, which is a vendor average and not an evidence record. For comparison, AiDASH reports USD 2M in efficiencies in National Grid's first few years and USD 1M in avoided cost on about 13,500 line miles, without saying whether the two overlap. |
What it leaves out: Counts spend only. It leaves out imagery and software costs, the reliability value of fewer tree events and customer minutes interrupted (which AiDASH reports for National Grid's worked circuits) or of beating reliability targets (which AiDASH reports for Entergy), avoided storm restoration cost and reduced wildfire risk.
Who already uses it?
5 public deployments, strongest evidence first. Grades: A regulator or audit, B the organization itself, C vendor case study, D anonymous or estimate.
Bear Valley Electric Service
United States · Energy and utilities · 2024
Bear Valley Electric Service (BVES) operates entirely within a High Fire Threat District in Southern California, where vegetation is the leading cause of outages and ignition risk. In March 2024 BVES deployed AiDASH's Intelligent Vegetation Management System (IVMS), combining satellite data with lidar, drone and ground inspections to map, monitor and prioritise vegetation risk.
AiDASH's customer story reports that BVES averaged about seven vegetation related outages a year between 2020 and 2024, and that since adopting IVMS in March 2024 the utility has recorded zero vegetation related outages and zero ignitions during wildfire season. Jared Hennen, BVES's Wildfire Mitigation and Reliability Engineer, says the utility has had no weather related outage since it started using AiDASH in March 2024, and that exporting clean, consistent data makes regulatory reporting easier.
No outcome disclosed.
PNM
United States · Energy and utilities · 2024
PNM, New Mexico's largest electricity provider, had High Fire-Risk Areas spanning up to 300 square miles. In 2024 it deployed AiDASH's CRIS satellite tool to deliver six hourly, recloser level wildfire risk insights from vegetation, weather and asset data.
Thad Petzold, PNM's Associate Director of Wildfire Risk and Vegetation Management, says the tool alerts the utility as it approaches operational thresholds so it can adjust system settings with confidence, and that the data helps justify PNM's risk based decisions to regulators, underwriters and investors. PNM was already an AiDASH client for satellite imagery, distance to centerline and hazard tree identification before adopting CRIS.
AiDASH states that the tool lets PNM reduce the size of Public Safety Power Shutoff zones, but gives no percentage or outage figure.
No outcome disclosed.
Trinity Public Utilities District
United States · Energy and utilities · 2023
Trinity Public Utilities District (PUD), a small utility in a wildfire prone, heavily forested part of Northern California, ran a three year vegetation management cycle on paper based plans. Prolonged drought and wildfire risk forced it to prioritise hazard tree removals over cycle trims, and despite removing more than 30,000 trees a backlog persisted.
In 2023 Trinity PUD implemented AiDASH's Intelligent Vegetation Management System (IVMS), starting with the hazard tree module and work planning, with mobile enabled field tracking and digital work plans replacing paper maps. AiDASH's case study reports that the utility is transitioning to a digitally trackable, field validated program and systematically reducing its backlog with data informed planning; it gives no outage, cost or backlog percentage figure.
Dave Delange of Trinity PUD is quoted saying the district is "building a more resilient, trackable vegetation program".
No outcome disclosed.
National Grid
United States · Energy and utilities · 2021
National Grid began working with AiDASH in 2020 in Massachusetts, a service area of over 13,500 line miles and more than 1.3 million customers. According to AiDASH, the utility had been on a five year trim cycle and had deferred work for four years running rather than fund it; in August 2020 it ran a proof of concept on its entire Massachusetts footprint, and the first model run produced its FY2021 work plan.
It adopted the Intelligent Vegetation Management System in 2021, which uses satellite imagery and AI to show vegetation conditions across the network, and moved to condition based trimming, with circuits now on cycles of four to seven years. AiDASH reports $1M in avoided cost from dropping manual field reviews and $2M in efficiencies in the first few years, without saying whether the two overlap.
It reports three sets of reliability results: on Massachusetts circuits worked in FY2022 to 2025, average improvements of 22% in tree events, 29% in customers impacted and 43% in customer minutes interrupted, measured 12 months after each circuit is worked; in its case study a 30% decline in tree related events, 38% in customers interrupted and 55% in customer minutes interrupted in the year after circuits were pruned; and from a talk by National Grid's vegetation strategy manager at the NextGrid Alliance Summit 2025, decreases of 26.4%, 30.2% and 46.5% on the same three measures.
- Cost savings: USD 2 million, in the first few years of adopting IVMS
"$2M in efficiencies realized – in the first few years of adopting IVMS."
Claimed by: vendor - Cost savings: USD 1 million, not stated
"$1M in avoided cost – with technology removing the need for time consuming manual processes like field reviews to determine if a circuit needs to be pruned."
Claimed by: vendor
Entergy
United States · Energy and utilities · 2020
Entergy, which serves more than 3 million customers through operating companies in Arkansas, Louisiana, Mississippi and Texas, uses the AiDASH Intelligent Vegetation Management System, which analyses satellite imagery with AI to show where vegetation threatens its power lines. According to AiDASH, Entergy moved from a standardized five year trim cycle to risk based maintenance, tailoring intervals so that some areas may need trimming every two to three years while others can go as long as thirteen years.
AiDASH reports that Entergy beat its vegetation reliability (Veg SAIFI) targets by over 30% in 2020 and 2021 on flat vegetation budgets, and quotes an Entergy vegetation management analyst saying all SAIFI targets were exceeded for all operating companies in the first year after IVMS was implemented and again in the second. A later AiDASH case study quotes Entergy's Vice President of Power Delivery Services saying vegetation impacts to customers improved by more than 20% within the first year.
No outcome disclosed.
How do you implement it?
A model agnostic playbook: what to prepare, the order to build in, and what goes wrong.
Data you need
- A GIS model of the network with spans, circuits and voltage
- Outage history per circuit with cause codes
- Trimming history and contractor work records
- Local knowledge of species, growth and regulatory clearance rules
Systems to integrate
- Geographic information system of the network
- Outage management system for cause coded outage history
- Work management system and contractor portals
- Mobile apps for crews and auditors
Complexity: Medium
Vendors deliver the imagery and models as a service. The work for the utility is a clean network model in GIS, outage history per circuit, and changing contracts and planning from fixed cycles to risk based work.
- 1
Prove it on real territory first
Run the imagery and risk model on a large, representative area and compare its risk ranking with outage history and a field check before changing the plan. National Grid, according to AiDASH, ran its 2020 proof of concept on its entire Massachusetts footprint rather than a portion of it, and the first model run produced its FY2021 work plan.
- 2
Agree the clearance rules and risk appetite
Encode regulatory clearance requirements and decide how much risk the utility accepts per circuit type, so the model plans to rules, not just to growth.
- 3
Move the plan from cycles to risk
Schedule circuits by predicted risk, keep a floor of mandatory inspections, and give planners the final say on the plan.
- 4
Change the contracts
Contractor agreements built on miles trimmed per cycle need to change to work orders by span and risk, with imagery based audits.
- 5
Measure reliability per circuit
Track tree related events, customers interrupted and minutes interrupted on treated circuits against the previous cycle and against untreated comparable circuits.
Guardrails
- Regulatory clearance and inspection obligations always override the model's schedule
- Planners approve the annual plan and any deferral of work on a high risk circuit
- Field verification of high risk findings before removal of trees on private land
- Imagery of private property used only for network maintenance purposes
KPIs to instrument
- Tree related outage events per 100 line miles, before and after
- Customers interrupted and customer minutes interrupted from tree causes (SAIFI and SAIDI contribution)
- Vegetation spend per line mile
- Share of high risk spans treated before the storm or fire season
- Audit pass rate of contractor work checked against new imagery
Human in the loop
Vegetation planners and arborists review the model's risk ranking, approve the work plan and decide on hazard tree removals. Field crews confirm conditions on site and report back, and reliability engineers review outcomes per circuit each year.
Common failure modes
- Model trusted over the rulebook
- A span with low predicted risk still has a legal clearance obligation. Keep regulatory rules as hard constraints.
- Stale imagery
- Imagery that is too old misses storm damage and fast growth. Agree refresh frequency by region and season.
- No change in contracts
- Contractors paid per mile on a cycle keep trimming on the cycle. Align contracts with risk based work orders.
- Benefits that cannot be shown
- Reliability varies with weather, so one good year proves little. Compare treated and comparable untreated circuits over several years.
What are the risks and rules?
EU AI Act
Depends on design
Annex III point 2 makes AI systems high risk when they are intended as safety components in the management and operation of the supply of electricity. Recital 55 defines such components as systems used to directly protect the physical integrity of critical infrastructure or the health and safety of persons and property.
A system that only feeds a multi year trimming plan, which vegetation planners review and approve before crews act, informs maintenance rather than directly protecting the network, and is then usually minimal risk. The assessment changes when the design acts directly on protection, for example when vegetation risk scores automatically trigger fire risk protection settings or switch lines off without a person deciding; such a system should be assessed as a possible safety component.
Standard GDPR duties apply where imagery shows private property or people.
Rules that apply
Guidance
- FAC-003-5 Transmission Vegetation Management (North American Electric Reliability Corporation (NERC), North America). The mandatory reliability standard for vegetation clearances on North American transmission lines, which any AI based plan must still meet.
- Annex III, high risk AI systems referred to in Article 6(2) (European Union, Europe). Point 2 lists safety components in the management and operation of critical infrastructure, including electricity supply.
- Recital 55, safety components of critical infrastructure (European Union, Europe). Defines safety components as systems that directly protect the physical integrity of critical infrastructure or the health and safety of persons and property, the test that decides the tier here.
Controls to put in place
- Documented mapping of regulatory clearance rules into the planning model
- Annual review of model performance against tree related outages per circuit
- Records of approvals for deferred work on high risk spans
- Data protection rules for imagery of private land
Frequently asked questions
- What results do utilities report from AI vegetation management?
AiDASH reports three sets of figures for National Grid.
Its latest account of the Massachusetts program gives average improvements of 22% in tree events, 29% in customers impacted and 43% in customer minutes interrupted on circuits worked in FY2022 to 2025, measured 12 months after each circuit is worked; a case study on the Massachusetts service area reports declines of 30%, 38% and 55% on the same measures in the year after pruning; and its page on a NextGrid Alliance Summit 2025 talk by National Grid's vegetation strategy manager lists decreases of 26.4%, 30.2% and 46.5%, without naming a service area.
The pages do not say which period the improvements are compared against, so treat them as indicative. For Entergy, AiDASH reports that it beat its vegetation reliability (Veg SAIFI) targets by over 30% in 2020 and 2021 on flat budgets, and quotes an Entergy vice president saying vegetation impacts to customers improved by more than 20% within the first year.
- Does satellite imagery replace field patrols and lidar?
Not fully. AiDASH's National Grid case study describes satellite imagery as a quick way to see vegetation conditions over an entire service area and a foundation for building a plan, and its later account notes that National Grid now also has lidar data and drone data from three in house drone pilots alongside the satellite outputs. Our advice: keep lidar, drones or field checks for precise clearance measurement and for verifying high risk findings before work on private land.
- Is AI vegetation management regulated as high risk AI?
Usually not under the EU AI Act when it only informs a maintenance plan that planners approve, because it does not directly protect the network in the sense of Recital 55. A design in which the risk scores directly trigger protective actions, such as fire risk settings or switching lines off, needs a proper assessment as a possible safety component. Existing clearance rules, such as NERC FAC-003 for North American transmission lines, still apply to the plan it produces.
How to cite this page
Blits.ai AI Use Case Library, "AI vegetation management for power lines", last verified 27 September 2026, https://www.blits.ai/ai-use-cases/power-line-vegetation-management. Licensed under CC BY 4.0. Method: how we verify use cases.
Changelog
- 29 September 2026: Added deployments: PNM, Bear Valley Electric Service, Trinity Public Utilities District.
- 27 September 2026: First published