Starting additional power plants to increase generation until the demand level is reached.

Sending a signal so that participating consumers reduce their load for the duration of the stress period.

Disconnecting residential customers automatically to preserve service for industrial sites.

Raising wholesale electricity prices until enough new generation becomes profitable to dispatch.

On-site hardware consumes less electricity than an equivalent cloud-based deployment of the same functionality.

On-site servers produce faster dashboard updates for building occupants who browse the BMS interface directly.

Cloud servers are unable to translate between the BACnet, Modbus, and KNX protocols that a BMS commonly uses.,m,m

Life-safety and continuous control functions must keep running even when the internet connection is lost.

Measure the electrical consumption of individual pieces of equipment inside a building and send the data to the utility.

Carry demand response event signals between the utility, the aggregator, and the participating building systems.

Directly control HVAC actuators, dampers, and lighting fixtures on the field bus inside the building.

Calculate the baseline consumption that is used to verify the size of a reduction after an event has ended.

Replace both OpenADR and BACnet with a single proprietary vendor protocol that unifies the two sides of the system.

Receive OpenADR events and translate them into BACnet commands that adjust setpoints, dim lights, and cycle equipment.

Store a baseline of past consumption on the BACnet side and transmit that baseline over OpenADR before every single event.

Encrypt the OpenADR traffic before it reaches the BACnet equipment so that the field bus remains secure from external access.

Real-Time Pricing, because the rate reflects wholesale market conditions and changes every hour of the day.

Critical Peak Pricing, because a very high rate applies on a limited number of stress days with day-ahead notice.

Interruptible Load pricing, because customers agree to reduce load by a fixed amount when asked by the utility.

Time-of-Use pricing, because the rate is set in advance on a daily schedule that customers can plan around.

Store all sensor data locally on the edge server permanently and eliminate cloud connectivity entirely

Reduce the CO₂ reporting interval to every 15 minutes so that less data is affected during an internet outage

Send all sensor data directly to the cloud and rely on the cloud to check CO₂ thresholds, because Wi-Fi is generally reliable

Process CO₂ threshold checks at a local edge server that can issue alerts independently of internet connectivity, while sending all other sensor data to the cloud for storage and trend analysis

The number of sensors sharing the gateway cost versus individual satellite fees, and whether the farthest sensor falls within LoRaWAN range given the coastal terrain

The brand compatibility between the sensors and the cloud analytics platform

The colour coding requirements for different monitoring zones and the preferred dashboard layout

The programming language used by the analytics platform and the data format of satellite transmissions

The sensors detect leaks in the irrigation infrastructure that the timer-based system could not identify

The sensors reduce the water pressure in the irrigation pipes, resulting in lower flow rates

The sensor-driven system eliminates watering of zones where soil moisture is already above the threshold, avoiding irrigation that the timer-based system applied regardless of actual conditions

The sensor-driven system uses a different type of water that requires less volume per irrigation cycle

The edge design continues checking thresholds and issuing alerts even during an internet outage, while the cloud-only design would fail to detect exceedances during connectivity loss, creating a compliance gap

The edge design uses more accurate threshold values because they are configured locally

The edge design produces alerts faster, and speed is the only factor that matters in regulatory compliance

The edge design costs less to operate because it transmits less data to the cloud

More frequent recalibration is unnecessary because 6% drift rate is within acceptable limits for all regulatory purposes

More frequent recalibration would improve data quality by catching drift earlier, but would approximately double the maintenance cost and effort for the lower-cost sensor fleet

More frequent recalibration would increase detection latency because sensors are offline during calibration

More frequent recalibration would reduce the lifespan of the sensors because the calibration process damages the sensing element