Managing the IoT….

IoT is finding its ground across industries….

Internet of Thing (IoT) has shown tremendous potential in its initial phase, and the tangible value generated has led to continued yet cautious enthusiasm in industry about this technology. During initial phases the industry has faced many challenges ranging non-availability of commercially viable sensors and gateways to no standard scalable platforms. However over the years with continued resilience the industry has overcome these challenges to some extent. Further the industry has moved beyond Proof of Concept (POC) & Proof of Value (PoV) to enterprise scale deployments.

 What has been IoTized…. How big is the challenge

Till date the industry has focused on adding IoT capabilities to the products and making them smart, leading to organic and sustainable development of IoT ecosystem.

Examples of same can be traced to Welbilt a Commercial Kitchen Equipment manufacturers connecting their kitchen equipment to cloud, or the other common examples that we see in day to day life are MG Motor's Hector providing advanced connectivity feature or Rotomatic a Smart Roti maker.

 According to an estimate there will be more than 1 Trillion connected devices by 2025 (World Economic Forum). On one hand this provides goose bumps to the IoT enthusiasts, whereas on other hand a reality check tells that managing such large amount of interconnected devices and data produced by them is not going to be a task of an order traditional IT Managed services have handled.

 Wait….But what has changed?

While traditional IT managed services model has been able to scale well for large scale IT deployments in enterprises, however with IoT the boundaries of enterprise has stretched beyond organization's premises. IoT devices are usually distributed across wide geographical area. Also, many times devices are placed at non-accessible places in environment that are potentially hazardous for humans. Hence organizations are required to be agile on not only delivering product innovations but also to ensure products and infrastructure availability to support SLAs. In this ecosystem OEMs are moving towards service orientation rather than limiting themselves to selling only the products to their customers. This essentially means OEMs now have to ensure their products are working and also underlying infrastructure is also working. For example Signify, a subsidiary of Philips and leader in smart lighting is providing their smart lighting solution to their customers and ensures service uptime and field support. But these developments create unprecedented challenges for OEMs which they might not be able to handle due to sheer volume of the problem. Managing such IoT deployments and ensuring SLAs are met requires adopting an innovative approach.

 IoT + RPA + AI = Being Future Ready

 IoT is about with acquiring real-time information. This information can be around environment surroundings or data from any sensor based on certain events. However, the real value from IoT can be generated when every information captured is analyzed and acted upon. This can be time-consuming and tedious task for an operator to keep track of this continuous stream of data. Robotic Process Automation (RPA), can help in managing such repetitive tasks and free up operators from monitoring this data.

 Robotic process automation (RPA) is the use of software bots to handle high-volume, repeatable tasks that previously required humans to perform. As IoT generates large volume of data, these bots can be configured to analyze this data, while operator intervention can only be requested for exception scenarios. An example of such scenario can be where connected vehicles are generating number of alerts, alarms and notifications. Only some of these alerts might require some action to be taken as per predefined standard operating procedures (SOP), like creating a spare part order or calling road side assistance, that can be implemented by RPA tools. Further, the health status of IT infrastructure to support IoT deployments can be continuously monitored through software bots. However, while managing IT infrastructure or application alerts there might be scenarios that are not part of SOP, and would intervention.

 In those scenarios cognitive technology or AI can support managing IoT landscape. While RPA aims at mimicking human actions, AI mimics the human thinking and reasoning process. AI enabled algorithms learns based on human actions without defining the scenarios explicitly. Such enabled AI system can take decision itself based on the scenarios or assist human operators with necessary information to enable him to take action quickly. A fully automated system that is capable of making decisions without human intervention would come closer to self-healing system.

 While on one hand, there is tremendous value that can be realized by leveraging RPA and AI to manage large scale IoT deployments, on the other hand technologies like AI are still in infancy. However, some initial steps can be taken by leveraging mature RPA solutions that can help in making large scale IoT deployments closer to self-managing scalable systems.