Large load growth, rising interconnection volumes, and increasingly complex planning requirements are changing the demands on Arizona's electric utilities. The question is whether grid data is ready to support that change.
Arizona's load growth is no longer just a long-term forecast. Arizona Public Service (APS) customers set a new all-time peak demand record of 9,164 MW on August 2, 2026, surpassing the previous record of 9,053 MW set just nine days earlier.
The longer-term outlook points to significantly more growth. APS expects peak customer demand to exceed 13,000 MW by 2038, approximately 60% above today's peak demand. APS identifies growth among commercial and industrial customers, including semiconductor manufacturing and expanding data center operations, as important contributors to rising electricity demand.
Salt River Project (SRP) is preparing for similarly significant growth. The utility expects its customer peak demand to increase by 50% by 2035.
Large commercial and industrial loads are contributing to that growth. Data centers are particularly significant in the Phoenix area. SRP reports that some larger data centers on its system have capacity loads of around 200 MW. Data centers contributed 5.1% of SRP's summer 2025 peak demand record, and SRP projects them to be its fastest-growing customer segment.
Much of this growth is also highly localized. A new data center, manufacturing facility, residential development, or charging site can significantly change loading conditions around individual substations and feeders. SRP notes that accurate load estimates from data center customers are necessary to appropriately size generation, transmission, and distribution infrastructure and plan for future demand.
As new loads and generation seek interconnection, Arizona utilities are under pressure to perform more engineering analyses in less time. That depends on grid data that can be integrated, validated, and maintained in a computable model.
When Grid Data Becomes a Bottleneck
Utility data is typically distributed across GIS, asset management, SCADA, MDM, engineering tools, interconnection portals, and other systems.
When this information is combined for engineering analysis, inconsistencies can emerge:
- Transformer ratings differ between systems
- Approved but not yet energized projects are missing from planning models
- DER records are incomplete or duplicated
- Switching states are inconsistent
- Asset parameters are outdated
These issues can be difficult to identify within individual source systems. A dataset may appear correct in isolation while producing implausible results once combined with topology, asset, load, or measurement data. Bringing this information together in a computable grid model allows utilities to identify inconsistencies through automated topology checks, load-flow calculations, and other validation methods.
The impact becomes particularly clear in interconnection studies. At Syna, grid compatibility testing for larger sites previously took up to eight hours. Following implementation of the envelio Intelligent Grid Platform (IGP), average processing time for interconnection requests was reduced from several hours to as little as 10 to 15 minutes.
A key part of this approach is maintaining a grid model that reflects assets already in operation as well as projects that have been approved but are not yet energized. Without these projects in the model, available capacity can be overstated and subsequent interconnection evaluations may be based on an incomplete view of expected grid conditions.
The envelio Intelligent Grid Platform provides a common engineering foundation for these workflows. Grid Transparency consolidates relevant data into a computable network model and automatically evaluates grid conditions. That model can then support downstream processes such as automated interconnection studies, queue management, and capacity visibility for prospective projects. Applications including Grid Connection Study and Grid Connection Navigator extend that foundation into formal interconnection evaluation and earlier-stage siting decisions.
Better Data Supports Better Planning
Data discrepancies can also influence decisions about where and when utilities invest in their distribution systems.
At FairNetz, calculations indicated unexpectedly high loading in one part of the grid even though station measurements showed relatively low demand. By combining information from different source systems in the grid model, FairNetz traced the discrepancy to storage-heating capacity recorded in its network information system that no longer reflected actual customer consumption. Further investigation ultimately allowed FairNetz to remove 25 MW of outdated storage-heating capacity from its NIS, equivalent to approximately 6% of the relevant substation capacity.
Without the correction, the inaccurate data would have continued to influence grid calculations and potentially the planning decisions based on them.
This becomes particularly relevant when load growth is geographically concentrated, as it is in parts of Arizona. New housing, manufacturing facilities, data centers, EV charging, and DERs can create very different requirements across feeders and substations.
For utilities deciding where and when to reinforce the distribution system, the accuracy of the underlying grid model directly influences the quality of those decisions. A model that overstates load can point toward investments that are not yet necessary. A model that understates load or fails to account for queued projects can obscure emerging constraints.
Improving grid data quality also enables utilities to evaluate flexibility as part of the planning process, rather than treating it only as an operational response after constraints appear. With a validated, time-aware grid model, planners can better understand where flexible interconnection, demand response, storage, or targeted capital upgrades will have the greatest system impact. This helps utilities avoid overbuilding for peak-based assumptions and make reinforcement decisions that support both near-term capacity needs and longer-term grid flexibility.
Improving grid data quality therefore provides a more consistent and defensible engineering foundation for distribution system planning, hosting capacity analysis, and interconnection decisions.
How Ready is Your Grid Data?
Arizona utilities are preparing for significantly higher electricity demand while managing new types of loads and increasingly complex requirements across their distribution systems.
Scaling engineering processes starts with knowing whether the underlying data is ready to support them.
If your engineering teams are spending more time preparing and reconciling models than running studies, it may be time to evaluate your grid data foundation.
Download our Grid Data Quality Checklist below to assess whether your current data environment can support faster technical review, improved model quality, and more transparent hosting capacity management.