Meter data management system: benefits and requirements

A meter data management system is the software that performs long-term data storage and management for the vast quantities of data delivered by smart metering systems: it imports reads from the head-end, validates, cleanses and processes them, and makes them available for billing and analysis. The primer on meter data management basics covers what the system is and how validation, estimation and editing work. This article is for the team about to specify or buy one: what the benefits are and what evidence supports them, what belongs in the requirements, how to size it, and what the organizations downstream of it actually see.

The scale is what makes the purchase unavoidable. In 2022, US electric utilities had about 119 million advanced metering installations, about 72 percent of all electric meters; residential customers accounted for about 88 percent of them, with roughly 13.9 million commercial and 575,000 industrial installations. Every one of those meters produces interval reads that a legacy billing system was never designed to hold.

Benefits, with the evidence behind them

The most complete public evidence comes from the US Department of Energy's Smart Grid Investment Grant program, which co-funded advanced metering at dozens of utilities and reported results in 2016. The program's 70 projects deployed more than 16.3 million smart meters, about 29 percent of those installed nationwide by 2014, and the report separates what the meters did from what the data systems behind them did. The benefits below are the ones a meter data management system is responsible for.

Billing accuracy and fewer estimated reads

Program utilities improved billing accuracy, reduced customer complaints and used AMI data to resolve billing disputes faster, and could notify customers of unusual usage before the bill arrived. That is the validation, estimation and editing engine at work: a read that fails validation is estimated, flagged and replaced when an actual read lands, instead of becoming a wrong bill. One vendor's benchmark is that a well-tuned system resolves 90 to 94 percent of reads on the first pass, leaving staff the genuine exceptions.

Operating cost

Nineteen reporting projects saved 316 million dollars in operations and maintenance cost over three years, an average of 16.6 million dollars per project, and program utilities avoided 13.7 million meter-operations truck rolls between 2011 and 2014. The average per-meter operations and maintenance saving was 8.37 dollars over a six-month period in 2014, with wide variation by project, and one rural cooperative in the program cut its annual meter operations cost by 65 percent.

Theft, tamper and outage detection

Tamper detection at one large Texas utility prevented revenue losses exceeding 450,000 dollars in 2012 and 130,000 dollars in 2014. On outages, one municipal utility in the program routes last-gasp alerts from its meters to its outage management system within two minutes, where they update the outage map. The meter data store is where that event stream is kept and correlated with accounts and locations, which is what makes a pattern of tamper flags on one feeder visible.

Settlement, forecasting and customer data access

Interval reads are also the basis for wholesale settlement, load research and rate design, and increasingly for handing data back to the customer. Green Button, based on the ESPI standard released by NAESB in the fall of 2011, is offered by utilities in nearly all 50 US states and two Canadian provinces, with Connect My Data feeds in 5-minute, hourly, daily or monthly intervals depending on what the provider makes available. In Ontario, most regulated electricity and natural gas utilities were required to provide Green Button access no later than November 1, 2023. The data a utility publishes through Green Button is the data in its meter data store; the feed cannot be cleaner than the store.

A requirements checklist

Requirements that every vendor can answer yes to are useless. The list below is written so each item can be tested on your own data during selection.

Sizing: how many reads you are actually storing

Sizing starts with one number: reads per meter per year at the interval you bill and analyze on. The arithmetic is fixed by the calendar.

Multiply by meters and by channels. A 25,000-meter utility on 15-minute intervals generates approximately 876 million reads per year on a single channel; a 100,000-meter utility on the same interval stores about 3.5 billion, and a second channel for received energy or reactive power doubles it. Add the event stream: outage alerts, tamper flags, voltage excursions. Then multiply by the retention your regulator and your load-research team expect, which is typically several years. The same vendor puts a mid-market implementation at 20 to 24 weeks from contract signing to first validated billing cycle; migrating history is usually the long pole.

Program utilities in the DOE study reported meters measuring consumption on 5-, 15-, 30- or 60-minute intervals, and 42 percent integrated AMI with billing, CIS and outage management, while 33 percent integrated with billing and CIS only. The interval you choose and the systems you connect drive size and cost more than any feature on the list.

The customer-side view: organizations that consume MDM outputs

Every commercial customer, energy consultant and reporting team is a downstream consumer of some utility's meter data management system, whether they know it or not. They receive its outputs in two forms: the monthly bill, whose determinants came out of the VEE engine, and Green Button files or feeds, which are the interval store exported in a standard format. What they see when the system is weak is estimated reads that persist for months, gaps in interval history, catch-up bills that reverse and rebill, and unit or multiplier mismatches between the bill and the interval file.

For those organizations the practical work is the mirror image of the utility's: collect the bill and the interval data, validate one against the other, and normalize across dozens of utilities that each run a different system with different rules. The Green Button Connect My Data guide covers the authorization and feed side, modernizing a utility CIS covers the billing system the meter data feeds, and meter-to-cash billing accuracy covers where the errors enter.

How MartinAI fits

MartinAI sits on the consumer side of every meter data management system. Its utility data collection pulls Green Button Download My Data and Connect My Data, interval feeds and bill uploads from any utility, reads bills of any layout across every commodity, and validates the bill against the interval data and the tariff. For a utility, that is an independent check on what its customers are receiving; for a portfolio, it is the one clean record that energy management, benchmarking and finance can share.