Page 36 - RC19 RealcommEDGE 2019 Fall Issue

Business Solutions

ARE YOUR BUILDINGS IN THE BEST OF

HEALTH? CONTEXTUAL AI CAN TELL YOU

ACROSS THE GLOBE, humans have shifted focus from fail, What component(s) in the equipment might fail, and
illness management to wellness management. Gone Why it/they might fail.
are the reactive methods of waiting to fall sick. Today,
people prefer staying updated on their health parameters An AI model for compliance should normalize for com-
through wearable gadgets and applications that track pletely unhealthy assets, but should cater to partially
physical activity, food habits and more. In the future, we healthy assets while designing a constraint for efficient
might be able to see our comprehensive health status on usage—since its objective would be to predict the best
demand. Data models could start predicting when and operational condition to achieve compliance.
why we might face a particular health issue, using multi-
ple parameters such as food intake, sleep cycle, activity, To improve efficiency, two types of AI models can be
medical history, family history, age, gender, race, local considered. The first type involves improving the mea-
weather, altitude and more. surements that control the equipment to run efficiently
while achieving its compliance objective; e.g., forecasting
As we speak, this paradigm shift is already underway for the calibration needs of important sensors like outside
building and equipment operations. Your artificial intelli- air temperature sensors, as they are impacted by roof
gence (AI) strategy can create data models that deliver var- radiation differently over a full day. The other type involves
ious equipment objectives by utilizing their unique context predicting the most optimized configuration to get the
and reading their ‘symptoms’ correctly through a predictive optimum balance between compliance and efficiency.
platform. You can take proactive action to maintain or
improve health, instead of reacting when failure occurs. When dealing with a single site, AI/ML models can learn
from the characteristics of the site. For most enterprises,
The main objective of any equipment in a building however, multiple sites or multiple equipment of the same
is ensuring compliance and comfort. The business type are involved, e.g., a large pizza chain with restaurants
processes around such equipment help the building across the country, or a large retail chain with similar roof-
maintain good health and operate with maximum effi- top units on all its stores. In such cases, the model design
ciency at the lowest cost. Ideally, equipment data should should recognize the commonality across sites and bene-
be normalized separately for each of these objectives— fit from the experience of millions of pieces of equipment,
despite the data being the same. A three-pronged AI but also include a component which recognizes the dif-
strategy of objective categorization, ecosystem learning, ferences specific to the individual sites’ characteristics.
and simplified in-process application can help transform Thus, an intelligently designed AI/ML model learns from
reactive processes for comfort and efficiency into a both—the shared learnings across hundreds of sites, and
proactive process. This delivers multiple benefits and the unique learning from each site.
reduces the overall cost of transformation.
While the above design helps create an approach for
Different objectives would need different categories of different models that have different objectives, success
models; however, they can learn from each other. An of the AI or ML models is primarily dependent on the
AI/Machine Learning (ML) model for equipment health cost-effective application and operationalization. Two
should be able to predict the four Ws of equipment fail- strategies for this are defined as the ecosystem learning
ure—Whether the equipment is going to fail, When it might and the simplified in-process application.

34