Building energy management system vs BMS vs EMIS: who needs what

Three acronyms get used interchangeably in RFPs and rarely mean the same thing. A building automation or management system controls equipment. A building energy management system controls equipment and monitors energy. An energy management information system takes meter data, bills and weather across a whole portfolio and turns them into baselines, benchmarks and reports. Buying the wrong one is expensive in both directions: an EMIS cannot start a chiller, and a BMS cannot tell finance why the gas bill rose.

The definitions below are drawn from vendor-neutral sources: the US Department of Energy, the national laboratory that runs its building analytics research, Natural Resources Canada, and the professional society that maintains the controls protocol most of these systems speak. The short definition of an EMIS is in what is an EMIS; this article is the comparison across all three layers, with a decision table at the end.

Layer 1: the building automation or management system (BAS or BMS)

A building management system is an integrated, computerised system used to monitor and control a wide range of building systems, which might include fire, smoke detection and alarms, motion detectors, CCTV, security and access control, lifts and so on, as well as systems such as lighting and HVAC. In North America the same thing is usually called a building automation system. Its job is control: schedules, setpoints, sequences and alarms.

Most of these systems communicate over BACnet, ASHRAE Standard 135, which the society describes as designed for heating, ventilating and air-conditioning control, fire and life safety, energy management, lighting control, physical access control and lift monitoring. The standard conveys building automation data between devices: hardware inputs and outputs, software values, schedules, alarms, logs and control logic.

Data produced: points. Thousands of them per building, sampled every few seconds to minutes: supply air temperatures, valve positions, fan status, zone setpoints, alarms. Trend logs of those points are the raw material for fault detection. Where it falls short: a BAS knows nothing about cost, tariffs, bills, weather-normalized performance or the building next door. It records what happened and does not diagnose why, and its trend storage is usually short.

Layer 2: the building energy management system (BEMS)

A building energy management system is an integrated, computerised system used to monitor and control specific energy-related building services plant and equipment, which will typically include HVAC systems, lighting and power systems. The same source notes that the two terms are often used as if interchangeable although their meanings differ: a BMS covers the broader set of building systems including security and fire, while a BEMS is focused on the energy-consuming ones.

In practice, a BEMS is a BAS with energy monitoring attached: main and submeter readings pulled into the same head end as the control points, energy dashboards for the building, and control strategies such as optimum start, demand limiting and night setback that are tuned on those readings. Data produced: everything the BAS produces, plus metered energy at the building or system level, at whatever interval the meters support. Where it falls short: it stops at the site boundary. It has no bills, no tariff, no view of thirty other buildings, and the energy figures it holds come from the building's own meters, which may or may not reconcile to the utility's.

Layer 3: the energy management information system (EMIS)

The US Department of Energy defines EMIS as a broad and rapidly evolving family of software tools that monitor, analyze, and control building energy use and system performance, listing utility bill management, interval meter analytics, measurement and verification, automated fault detection and diagnostics, supervisory control and operations and maintenance optimization among its capabilities, and naming building automation systems, utility bills, weather data and advanced metering as its data sources. The BAS, in other words, is an input.

The national laboratory behind DOE's building analytics work splits the family into two main tools. An energy information system provides, at minimum, daily, hourly or sub-hourly interval meter data at the whole-building level, with graphical and analytical capability. Fault detection and diagnostics tools continuously analyze system-level operational data to find HVAC faults and their causes, and many FDD tools integrate the trend log data from a BAS but otherwise are stand-alone software packages. Natural Resources Canada's handbook characterizes an EMIS by its deliverables, early detection of poor performance, support for decision making and effective energy reporting, and sets the bar that a good one should reduce energy use and cost by at least 5 percent.

The savings evidence is the strongest of the three layers because it has been measured at scale. DOE's Smart Energy Analytics Campaign, covering 96 organizations, 518 million square feet and nearly 6,000 buildings, reported median energy savings of 4 percent (0.04 dollars per square foot) for energy information systems and 9 percent (0.24 dollars per square foot) for FDD, with a one to two year simple payback. The peer-reviewed analysis of the same campaign put median base cost to install an EMIS at 0.03 dollars per square foot, with annual software cost of 0.02 dollars and estimated annual labor of 0.03 dollars per square foot, and found FDD base cost at 0.05 dollars per square foot, five times the EIS figure.

Data produced: normalized consumption, cost and demand per site and meter, baselines and avoided energy, benchmarks, emissions and exception lists. Where it falls short: an EMIS does not control anything unless a supervisory-control module is added, and every result it produces inherits the quality of the bills and meter feeds underneath it.

How they connect: points, meters and bills

The three layers are joined by three kinds of data, and the direction of flow matters. Control commands flow down, from the BAS to the equipment, and only the BAS or BEMS sends them. Points flow up: BACnet trend logs from the BAS feed FDD tools and, in a BEMS, sit beside the meter readings. Meter data flows up from interval meters, submeters and Green Button feeds into the EIS layer of an EMIS. Bills flow in from the utility and land only in the EMIS, because neither of the lower layers has anywhere to put a tariff line.

• BAS to EMIS: a BACnet or export gateway that pushes trend logs on a schedule. The FDD module needs system-level points; the EIS needs only whole-building meters. Decide which before paying for the gateway.
• Meters to EMIS: utility interval data through Green Button Connect My Data, and submeter data from the BEMS or a separate metering network. The two should reconcile to the utility bill, and usually do not until someone checks.
• Bills to EMIS: extracted fields (consumption, demand, cost by line, period dates, rate class), validated against the meter data and the tariff before they feed baselines and benchmarks.
• EMIS back to operations: the exception, not the rule. An EMIS finding becomes a work order or a setpoint change made by a person in the BAS. Fully automated supervisory control exists, but it is a separate product decision.

Decision table: who needs what

Where MartinAI fits

MartinAI is the data layer under the EMIS: it reads utility bills of any layout across every commodity, validates them against tariffs and meter data, pulls Green Button and interval feeds, and delivers one clean record into its energy management boards, benchmarking, weather normalization, M&V and emissions workflows, and onward to ENERGY STAR Portfolio Manager and RETScreen. It does not replace a BAS or a BEMS; it makes the meter and bill side of the picture trustworthy so the analysis above them holds. The requirements for the analytics layer are in choosing an energy management information system, and what meter data adds to controls-based diagnostics is in fault detection and diagnostics with utility data and monitoring-based commissioning.