Advanced analytics applications in IoT

The new age of technology Internet of things is all over the place. The real benefit of IoT comes with inter-connecting more and more devices and our ability to harness this data to take quicker decisions leading to accurate and timely actions by leveraging Advanced Analytics.Advanced Analytics has become a key driving factor to define the success of a company irrespective of the business.

Imagine the days of no online shopping or the convenience of calling a cab via a mobile app on a rainy day or unable to check the weather for the day online or find the route to drive home. Internet of things has advanced in endless areas of our lives like manufacturing, healthcare, social media, e-commerce etc., more so in areas that impact our personal and social lives like weather, connected homes and soon connected cars and even connected elevators. After all, what will we do without our smart phones on our side with its array of sensors to wirelessly connect us to this world? Our smart phones have gotten smarter while we are getting dumber by being so reliant on them.

We have a huge gap when it comes to our capability to harness this potential. Businesses sit on top of piles of digital transformation data with no way to extract a strategic plan that works to be profitable. The challenges faced include how to effectively apply advanced analytics on this overwhelming data and also how to identify and prioritize their application areas.

IoT implementation in applications are far too many and impacts all areas starting from data inception, consumption to mining foresights from the data.Business intelligence systems and analytics systemsare used interchangeably these days because of the poor understanding of its true meaning.Objectives of business intelligence systems are to churn out management reports in the form of structured reports that can be consumed by management operations. Reports will also include Push or pull notifications like SMS, email etc. On the contrary, Analytics starts where business intelligence ends with a key distinguishing factor being complexity of calculations performed by leveraging mathematical based models and their outcomes.

A basic application of analytics in IoT where in there is a predetermined relationship between a set of parametric values are known to indicate a specific outcome of another
parameter within the system. These parametric values are nothing but data read from sensors mounted on various IoT subsystems like temperature, pressure etc. These rules are knowledge acquired by subject matter experts about the system which is translated and embedded as software code for raising an alarm when the conditions are met.

Anomaly detection is a step up where in using statistical techniques, a deviation from the normal operating ranges of an equipment is identified and alerts are raised. Anomaly can be due to a single parameter or a combination of parametric values. Next higher in the order of complexity is change pointdetection wherein a statistically driven algorithm detects a change in performance of the system. Change point detection is often confused for rule based deviation detection. The primary distinction is that change point detects any permanent departure in the performance of a system while a rule based change detection detects a point in time change which might fall back to normal sooner or later.

One of the advanced applications of analytics in IoT is Predicting systems behaviour. These can be either forecasting future state of a parameter using time series models and multivariate event driven predictive models. The target for prediction can be linear values like binary or conditions like yes or no. One of the most complex areas to implement advance analytics predictive model is real time control of automated processes in manufacturing like welding, spray painting etc.

While advanced features like edge analytics may sound glamorous from commitment to modernizing an organization perspective, it is recommended to consider a very detailed evaluation to justify the need for such disposition.The closer one moves towards the edge layer, more tighter coupling between the software and hardware platforms is needed. Thus, any dispositions on the edge require an overhaul of the hardware controlling the equipment, and hardwiring them to analytics capable software systems.To what extent a decision and an action translation can be allowed in an automated system without human intervention to judge validity of such machine generated decisions?