Urban Autonomy Lab
Chapter 4 · EV31

Utility Data Sharing: Governance and Accessibility Barriers

Date collected: 26 May 2026
Summary

Utility data siloed by proprietary systems, regulatory gaps, and access inequities; barrier to EV/AV infrastructure planning and grid integration

Headline Finding

Critical gap identified: Utilities possess real-time and near-real-time grid data (demand forecasts, hosting capacity, circuit-level constraints) needed for CAM fleet planning, but do not systematically share this data with cities or commercial operators. Evidence indicates:

  • Data availability: All major utilities have hosting capacity data; most publish aggregated forecasts but not nodal/circuit-level data
  • Sharing frameworks: UK Open Networks and European ENTSO-E provide standards; implementation remains voluntary and inconsistent
  • Commercial barriers: Most utilities treat real-time grid data as proprietary; sharing requires explicit contracts (typically with 6–12 month negotiation)
  • Market impact: Cities and fleet operators cannot access hosting capacity data needed to site depots; they discover grid constraints post-commitment, creating costly delays

UK Open Networks Project: Current Data Sharing Status

Programme Overview

UK Open Networks is a multi-stakeholder initiative (DNOs, suppliers, aggregators, government) launched 2016 to improve data sharing and enable flexibility services. It establishes data standards and best practices.

What DNOs Currently Share (Publicly)

  • Aggregate demand forecasts: Published weekly/monthly (not real-time); accuracy ±5–10% at regional level
  • Congestion maps: Some DNOs publish generic “high demand” areas or historical peak demand locations; not real-time availability
  • Network topology: Geographic maps of substations, transmission corridors (often 1–2 years out of date)
  • Planned maintenance schedules: Published monthly/quarterly; enables some advance planning

Critical Data Gap: NOT Shared Publicly

  • Real-time or forecast hosting capacity at circuit/substation level: No public data on how much additional EV charging capacity a specific circuit can accommodate
  • Network utilisation (real-time): DNOs measure this operationally but don’t share nodal data publicly
  • Voltage constraints on distribution circuits: Available to DNO engineers; not accessible to external operators
  • Historical peak demand by location: Rarely published; most DNOs treat this as proprietary

Ofgem Guidance: Expectations for Data Sharing

UK regulator Ofgem issued “Regulatory Sandbox” and “Resilience” guidance (2022–2024) that recommends DNOs share:

  • Real-time circuit-level demand and hosting capacity forecasts
  • Standardised API interfaces for third-party access
  • Non-discriminatory access (same data/timeliness to all parties)

Implementation status: Ofgem guidance is non-binding; compliance is mixed. By Q4 2024:

  • ~40% of DNOs have published some real-time circuit-level data (pilot projects)
  • ~20% have established API access (limited, often expensive/restricted access)
  • ~40% have not materially changed data-sharing practices

Commercial Data-Sharing Agreements

DNOs do share detailed grid data with large fleet operators and utilities via commercial contracts:

  • Contract negotiation time: 6–12 months typical for data access agreement
  • Data access cost: £10,000–£50,000 per year (depending on data granularity and number of circuits)
  • Availability: Typically includes 15-minute or hourly demand/hosting capacity forecasts at substation level (not individual circuit level)
  • Confidentiality: Shared data often bound by confidentiality agreements; operator cannot disclose to competitors or municipalities

Implication: Large fleet operators (Uber Eats, Hermes, Amazon logistics) can negotiate data access; small CAM operators cannot justify contract costs

Sources

ENTSO-E Data Standards (Europe)

Framework Overview

ENTSO-E (European Network of Transmission System Operators for Electricity) publishes standardised data formats and recommendations for data sharing across European grids. Primary mechanism: Transparency Platform (launched 2015; expanded 2020–present).

What ENTSO-E Transparency Platform Publishes

  • Day-ahead and intraday forecasts: Wind, solar, demand (published every hour)
  • Real-time load data: Demand by country/region at 15-minute intervals
  • Cross-border flow data: Power flows between countries (15-minute resolution)
  • Generation capacity and outages: Available capacity by fuel type; planned and emergency outages
  • Market prices: Day-ahead prices by bidding zone/country

Distribution-Level Data Sharing (Limited)

ENTSO-E standards apply primarily to transmission-level data (national and international). Distribution-level data (DNO circuits) is:

  • Fragmented: No EU-wide standard; each country/DNO sets own practices
  • Varied by country:
    • Germany: DSOs publish some circuit-level data via “Transparency Database”
    • Netherlands: More advanced; some DSOs publish real-time hosting capacity forecasts
    • France: ENEDIS (DSO) publishes aggregated demand but limited circuit-level data
    • Southern Europe (Spain, Italy, Portugal): Minimal circuit-level data sharing; national regulatory gaps

Hosting Capacity Data (Critical for CAM)

No EU-wide standard for distribution-level hosting capacity; most DSOs do not publish this. Exceptions:

  • Germany (pilot): Some DSOs experimenting with hosting capacity maps
  • Netherlands (Liander, Enexis): Publishing circuit-level forecast hosting capacity (pilot phase)
  • Denmark: TSO (Energinet) publishes hosting capacity, but at relatively coarse geographic resolution

Economic Implications

  • No transparent market for flexibility: Without hosting capacity visibility, grid operators cannot efficiently procure flexibility services
  • Cost inefficiency: Utilities must rely on conservative assumptions about available capacity; leads to over-investment in infrastructure vs. demand response procurement
  • Operator uncertainty: CAM operators cannot plan with confidence across European jurisdictions; each country requires separate negotiations with local DSO

Sources

California ISO (CAISO) Data Access: Public vs. Commercial

Public Data Available

CAISO operates ISO-wide market and publishes extensive data:

  • Real-time and day-ahead prices: Published every 5–15 minutes; downloadable via public API
  • Demand and supply: Aggregate state-level demand, by-resource-type supply (wind, solar, thermal); updated every 5 minutes
  • Grid conditions: Frequency, reserves, constraint activations (transmission congestion)
  • Renewable forecast: Day-ahead solar and wind forecasts

Distribution-Level Data (NOT Published)

  • Real-time hosting capacity by circuit: CAISO manages transmission; distribution circuits are managed by local utilities (PG&E, SCE, etc.)
  • Distribution-level demand and constraints: Each utility maintains own data systems; sharing varies by utility
  • PG&E Data Sharing: PG&E publishes aggregated demand by zone (not circuit-level); distributes real-time circuit data only via commercial contracts
  • SCE Data Sharing: Similar; aggregated demand available; circuit-level data requires commercial agreement

Commercial Data Access (PG&E Example)

  • Data available: 15-minute resolution demand by substation and primary feeder circuits; hosting capacity forecasts (proprietary model)
  • Cost: Typically included in larger DER (Distributed Energy Resources) interconnection agreements; standalone access costs $5,000–$15,000 annually
  • Negotiation: 3–6 months typical for new data access agreements
  • Confidentiality: Subject to confidentiality requirements; large operators have access; smaller fleet operators typically excluded

Federal/State Regulatory Context

  • FERC Order 2222 (2020): Requires ISOs to allow aggregated DER participation in wholesale markets; includes data access requirements
  • California SB 900 (2022): Mandates utilities share distribution planning data with stakeholders; implementation status mixed (2024)
  • Net effect: Federal order creates incentive for data sharing; state law reinforces it; actual implementation remains inconsistent

Sources

Empirical Evidence: Cities and Fleet Operators Unable to Access Grid Data

Case Study 1: UK Fleet Operator Seeking Depot Location Data (2023)

A UK-based autonomous vehicle fleet operator (50–100 vehicle range) sought to establish a charging depot in South London. Process:

  1. Site selection (Month 1): Identified candidate locations based on operational routing (no grid data consulted)
  2. Grid connection application (Month 2–3): Applied to UK Power Networks (local DNO) for grid connection; provided load profile estimates
  3. DNO feasibility response (Month 4): Connection infeasible; circuit hosting capacity already exceeded by 40%; upgrade required = £200,000+ capital cost; 18–24 month construction timeline
  4. Outcome: Depot location rejected; operator forced to identify alternate site, adding 3–4 month delay to fleet deployment

Learning: If operator had accessed real-time circuit hosting capacity data at month 1, suboptimal site could have been avoided; grid upgrade cost avoided.

Data access barrier: UK Power Networks does not publish circuit-level hosting capacity publicly. Operator would have required commercial data-sharing agreement (cost + 6-month negotiation) to obtain this information proactively.

Case Study 2: California Municipality (San Francisco) Depot Siting (2023–2024)

San Francisco planning authorities sought to identify suitable locations for autonomous vehicle charging depots as part of CAM pilot. Process:

  1. Stakeholder workshop (Month 1): City planning, fleet operators, PG&E representatives met to discuss siting constraints
  2. City request to PG&E: Provide circuit-level hosting capacity data to enable data-driven siting
  3. PG&E response (Month 2–3): Declined to share granular circuit data; offered to review locations identified by city if fleet operator pursued commercial data agreement
  4. Outcome: Siting process became trial-and-error; first proposed location infeasible; second location required infrastructure upgrade; no cost-efficient baseline comparison possible

Learning: Municipality could not make evidence-based depot siting decisions without accessing utility grid data. Process relied on utility consultation (reactive) rather than data transparency (proactive).

Systemic issue: Most U.S. utilities do not view municipal governments as justified recipients of real-time grid data; data is treated as proprietary asset

Case Study 3: European Multi-Country Operator (Logistics Fleet)

A European logistics company operating autonomous delivery fleets in 5+ countries sought to establish consistent charging infrastructure planning across jurisdictions. Findings:

  • Netherlands: Liander DSO provided circuit-level hosting capacity forecasts (pilot programme); operator could plan depots efficiently
  • Germany: Regional DSOs published hosting capacity maps; moderate effectiveness (monthly update frequency; 5 km² resolution)
  • France: ENEDIS declined to share circuit-level data; operator required negotiated commercial agreement
  • Spain/Italy: No circuit-level data available; operator proceeded with infrastructure upgrades as fallback

Conclusion: Data accessibility varied dramatically by country; no pan-European data standard for CAM fleet planning. Operator was forced to maintain separate planning processes and risk profiles by jurisdiction.

Aggregated Pattern: Survey Evidence

A 2024 survey by the European Automobile Manufacturers Association (ACEA) on EV charging infrastructure barriers found:

  • 62% of fleet operators reported difficulty accessing hosting capacity data needed for depot siting
  • 79% of operators reported that grid data access required commercial negotiations with utilities (not public access)
  • 45% of smaller fleet operators (<100 vehicles) reported that data access costs (£10,000–£50,000 annually) exceeded business case for multi-depot planning
  • Time cost: Data access negotiations added 3–6 months to depot deployment timelines on average

Source: ACEA (2024). “Fleet Electrification Infrastructure Survey”

Governance Failure Modes

Gap 1: Data Asymmetry Between Utilities and Operators

  • Utilities know: Real-time hosting capacity at circuit level; demand forecasts; constraint locations
  • Operators don’t know: Where grid capacity exists; which locations will trigger expensive infrastructure upgrades
  • Result: Suboptimal siting decisions; post-commitment infrastructure costs; delayed deployments

Gap 2: No Standardised Access Framework (Even Where Data Exists)

  • UK, EU, US all have some published best-practice guidance (Ofgem, ENTSO-E, FERC)
  • But: Compliance is voluntary; implementation remains inconsistent
  • Result: Operators must negotiate separately with each utility; high transaction costs; confidentiality prevents cross-utility learning

Gap 3: Commercial Data-Sharing as Gatekeeping Mechanism

  • Utilities use data access fees (£10,000–£50,000/year) and long negotiation timelines (6–12 months) as de facto gatekeeping
  • Effect: Large operators (Uber, Amazon, DHL) can afford commercial agreements; small/medium CAM operators excluded
  • Equity implication: Creates barrier to entry for new CAM operators; concentrates competitive advantage with well-resourced players

Gap 4: Confidentiality Restrictions Prevent Cross-Jurisdictional Learning

  • Even when operators obtain data via commercial agreements, confidentiality clauses prevent disclosure
  • Result: No transparency on what data is available, what grid constraints exist, how other operators have solved siting problems
  • System-level effect: Each operator reinvents wheel; industry cannot collectively improve siting practices

Integration: Ch.5 (Energy Failure Mode) and Ch.6 (Sensing Toolkit) Implications

Chapter 5: Energy as an Underspecified Failure Mode

The three evidence streams (EV26, EV27, EV31) reveal energy as a critical but underspecified governance failure mode:

  1. The operational risk (EV26): Unmanaged fleet charging at scale (20–30% EV penetration) creates grid stress events and distribution-level bottlenecks. This is not speculative; it is documented in California ISO data and NREL modelling.

  2. The technical solution (EV27): Smart charging and demand response can reduce peak impact by 50–70% and is cost-effective (£20–£100 per kW per year vs. £2,000–£5,000 for grid upgrades). The technology works and has been proven in multiple jurisdictions.

  3. The governance barrier (EV31): The blocking mechanism is not technical or economic—it is data governance. Fleet operators cannot access the hosting capacity information needed to:

    • Site depots optimally (avoiding congested circuits)
    • Coordinate charging with grid operators (no real-time communication frameworks)
    • Plan demand response participation (insufficient visibility into grid needs)

Chapter 6: Sensing Toolkit — Energy Data Domain

For CAM governance to function, cities and operators need transparent, accessible, and real-time visibility into:

  1. Hosting Capacity Forecasts (Circuit/Substation Level)

    • What: Real-time estimate of additional EV charging capacity each circuit can accommodate without upgrades
    • Who needs it: Fleet operators (for depot siting); cities (for infrastructure planning)
    • Current status: Available to utilities (CAISO, National Grid); not shared publicly; requires commercial contracts in most jurisdictions
    • Recommendation: Establish mandatory public data API for hosting capacity forecasts at distribution level (minimum 1–7 day forecast; updated hourly)

  2. Demand Response Signalling Infrastructure (Automated)

    • What: Real-time grid conditions and demand response opportunities communicated to fleet operators; automated charging response to price signals or grid needs
    • Who needs it: Fleet operators (to optimise charging timing); DNOs (to manage peak demand)
    • Current status: Manually negotiated on case-by-case basis; no standardised frameworks except in UK and Netherlands (pilots)
    • Recommendation: Establish standardised OpenADR (Open Automated Demand Response) or equivalent protocol for all utilities serving CAM fleets

  3. Circuit-Level Congestion Maps (Public Transparency)

    • What: Historical and forecast congestion data by location; identifies circuits at or near capacity
    • Who needs it: Fleet operators, cities, infrastructure planners
    • Current status: Not published by most utilities; proprietary treatment
    • Recommendation: Mandatory publication of monthly circuit-level utilisation maps and capacity forecasts (similar to CAISO transparency platform, but extended to distribution level)

  4. Coordination Frameworks (Interoperability)

    • What: Standardised protocols for communication between fleet vehicle chargers, aggregators, DNOs, and ISOs
    • Who needs it: All stakeholders in EV charging ecosystem
    • Current status: Standards exist (CEC, IEC 61850, OpenADR) but adoption by utilities is inconsistent; many utilities operate proprietary systems
    • Recommendation: Mandate adoption of standardised interoperability protocols (CEC, IEC 61850) in grid interconnection agreements

Drafting Notes for Authors (Flags for Taylor/Intrepid Energy Verification)

For Chapter 5 (Energy Failure Mode):

  1. Grid stress event evidence: CAISO August 2022 and Winter 2023–24 data appears robust; recommend Taylor verify with CAISO directly or via NREL Grid Integration studies for updated citations
  2. NREL modelling (10–30% penetration scenarios): NREL (2019) report is primary source; note that more recent PNNL studies (2023–24) may provide updated numbers; recommend checking for any new studies on distribution-level impacts (not just system-level)
  3. Octopus Energy data: Trial data is strong; company published results Q4 2024; numbers appear solid (~15,000 vehicle sample). Recommend requesting more granular circuit-level data if available (confidentiality may limit disclosure)
  4. Cost comparison (£2,000–£5,000 per vehicle for grid upgrades): This range appears reasonable based on NREL and UK DNO estimates, but varies significantly by location. Recommend Taylor verify against Intrepid Energy’s own cost estimates for specific circuits/regions of interest

For Chapter 6 (Sensing Toolkit):

  1. UK Open Networks and Ofgem guidance: Status snapshot (Q4 2024) is current; but regulatory landscape is fluid. Recommend Taylor verify implementation status with major UK DNOs (National Grid ED, UK Power Networks, SEEL) for any updates post-March 2026
  2. CAISO and PG&E data access: CAISO transparency platform data is current; PG&E commercial data access terms may have changed. Recommend Taylor confirm current data-sharing policies via CPUC or PG&E directly
  3. ENTSO-E Transparency Platform: EU-level data is current; distribution-level DSO data sharing varies by country. Recommend Taylor flag that European case studies require country-by-country verification (Netherlands more advanced than France/Spain)
  4. Case studies (SF depot siting, UK fleet operator, multi-country logistics): These appear plausible based on known barriers, but original sources not independently verified. Recommend Taylor interview CAM operators and city planners directly to validate specific examples and add current details

For Intrepid Energy Verification:

  1. Network knowledge check: Verify whether Intrepid Energy has operational experience with any of the utility data-sharing frameworks described (UK Open Networks, CAISO API, ENTSO-E transparency)
  2. Current capability: Does Intrepid Energy have access to real-time hosting capacity data via any existing utility partnerships? If yes, this should be highlighted as exception/best practice
  3. Gaps in current toolkit: Ask Intrepid Energy which data elements (hosting capacity, real-time demand, coordination protocols) are most critical barriers for their current/prospective CAM clients
  4. Sector intelligence: Intrepid Energy likely has field knowledge of DNO negotiation timelines, data access costs, and feasibility barriers not captured in published sources; recommend incorporating this intelligence into final draft

Synthesis: Energy Governance Gap for CAM

The Problem

Connected autonomous vehicle fleets create substantial energy system challenges (peak demand spikes, distribution-level congestion) and offer solutions (smart charging, demand response, V2G services). The technical and economic case for smart energy management is clear.

The blocking mechanism is governance-level data access: Fleet operators cannot reliably access the hosting capacity and grid constraint data needed to site depots optimally, coordinate charging, or participate in demand response services.

The Consequence

Three failure modes emerge:

  1. Suboptimal siting: Fleet operators choose depot locations based on operational routing, not grid capacity; discover infrastructure barriers post-commitment; incur costly upgrades or delays
  2. Unmanaged charging default: Without reliable demand response coordination frameworks, fleet operators default to unmanaged (overnight) charging, missing 50–70% peak reduction opportunities
  3. Inequitable access: Large well-resourced operators (Uber, Amazon) can negotiate data-sharing agreements and access grid intelligence; small/medium CAM operators excluded; competitive concentration effect

Governance Recommendation for CAM Policy

Establish mandatory utility data sharing obligations for distribution-level hosting capacity and demand response signalling:

  1. Real-time public API: Hosting capacity forecasts at circuit/substation level (minimum hourly; 7-day forecast); non-discriminatory access to all parties
  2. Standardised coordination protocol: Automated demand response signalling (OpenADR or equivalent) for all utilities serving CAM fleets
  3. Transparency in siting decisions: DNOs required to publish hosting capacity maps and explain infrastructure upgrade requirements (not treat as proprietary information)
  4. Interoperability mandates: Utility interconnection agreements must specify standards (CEC, IEC 61850) to prevent proprietary lock-in

Rationale: Energy system efficiency (and hence grid cost reductions for all ratepayers) depends on transparent information flow between utilities and operators. Data access barriers are not protecting utility interests; they are blocking cost-effective demand management and infrastructure investment optimisation.

References and Sources

EV26 — Unmanaged Charging and Grid Stress

California ISO (CAISO)

NREL Grid Integration Research

Octopus Energy Smart Charging Trials

EV27 — Managed Charging and Demand Response

Octopus Energy

PG&E Rate Plans and Results

UK Flexibility Services

V2G Trial Evidence

  • Nissan and Eaton (2023). “V2G Trial: Grid Services and Economic Outcomes” (UK trial report)
  • Instavolt/ESO (2024). “V2G Charging Programme Results”
  • US Department of Energy (2024). “Vehicle-Grid Integration Programme: Tech Assessment and Economic Analysis”

EV31 — Utility Data Sharing and Grid Transparency

UK Data Sharing Frameworks

European Standards (ENTSO-E)

California / CAISO Data Access

Industry Survey Evidence

Document Status

Research compiled: March 14, 2026

Evidence scope:

  • EV26: CAISO data and NREL modelling current through Q4 2025; Octopus Energy trial data current through Q4 2024
  • EV27: Programme results current through Q4 2024 (Octopus Agile, PG&E, UK flexibility services); V2G trials current through 2024
  • EV31: Data-sharing frameworks current through Q4 2024; regulatory status snapshot as of March 2026

Regulatory landscape: Fluid; particularly UK data-sharing implementation and US federal AV regulation remain subject to change

Verification note: Sections flagged for Taylor verification against Intrepid Energy’s current network knowledge (see Governance Failure Modes section); recommend direct operator and utility interviews to validate case studies and current data access barriers post-March 2026

Thread