Originally published on www.m2mnow.biz
The Internet of Things (IoT) has become a buzzword in recent times and especially in the past few years. However, the possibilities and the business merits of connected assets are now being understood much better.
IoT should just be seen as an enabler to a broader result – not an end in itself. Connectivity is commodity, says V. R. Vijay Anand of Wipro Ltd. The true value of IoT comes from the business benefits, across the ecosystem that can be realised by connecting the assets. It’s very much like a new gadget that has its ‘wow moment’ with all the cool functionalities. However, once the initial euphoria of the “cool functionality” has passed, we start looking for the value and the return on investments on the gadget.
IoT is very similar to this new gadget. Unless we are able to actually associate a quantifiable business value from the IoT investment, the success that we foresee and the volumes of connected assets predicted, will always be a moving target.
Business value comes from business insights and these insights are enabled by analytics of the machine data, mashed up with enterprise, social and environmental data. The value increases many fold if we are able to source data from across the breadth of seller, partner and customer ecosystems.
The success of any IoT initiative is also driven by the ability to create value across the entire stakeholder ecosystem of a business – be it the services organisation, the manufacturing plant, operations, sales, or R&D. The insights required to create value for each of these might be different and specific; but in all cases traditional enterprise data, when complimented with machine data, helps the enterprise move to the next level of maturity and differentiation in its operations.
Broadly, analytics can be categorised into four levels, which signify their level of maturity. Every level introduces a layer of complexity but at the same time the values manifested by these levels increases many fold.
- Descriptive analytics – this is the basic level where data is able to tell us or ‘describe’ what is going on. It is the representation of the facts without any analytical processing. For example, this might be a health care OEM being able to remotely monitor the status of all their equipment’s across the health care facility of their customer.
- Diagnostic analytics – at this level of maturity, mashing up relevant external data (i.e. data that is not machine generated), can allow you to start to diagnose the reasons for certain trends. For example the health care OEM can combine machine data and historical service data to understand the root cause of a certain service or technical problem.
- Predictive analytics – This is an advanced level of analytics where you can start to predict certain possibilities or occurrences based on current and historical trends. This can help business to plan their operations in advance and plan for any breakdowns or downtime. For example the health care OEM might be able to predict that a piece of equipment with one of its customers will have a failure in 3 weeks and proactively schedule a service, so that the down time of the equipment is completely avoided – or at worst minimised.
- Prescriptive analytics – Going beyond just predicting an occurrence, prescriptive analytics can actually prescribe a way to prevent the occurrence or even provide a solution to fix the problem and, as such, businesses can completely reinvent their process and models. For example the health care OEM is able to predict that one of its customer’s equipment is going to fail in the next 3 weeks. The manufacturer also gets the insight that the failure is because the equipment is being used incorrectly and is therefore able to go back with prescriptions on the right usage. Equally, if you know that machinery piece of equipment is going to break down during a certain period of time, along with the service engineer being despatched, you could predict the spare parts required based on the automatic solution prescription and arrange for that part to be automatically ordered and available at the right time.
While there is a fair amount of variation in maturity across different industries, the truth is that most enterprises are at maturity level 1 or 2 when it comes to deriving insights from their connected data. Many enterprises already have connected assets. However, while often being aware of the data being produced, many are often largely unaware or incapable of producing the kind of insight and solutions that proper analysis of this data can offer.
There is an immense opportunity as the IoT has the potential to address many of the issues that concern stakeholders right across the business – from an operations manager to a C-suite level executive, such as inefficiency, poor productivity, disruption in market places and increased competition. For example, a manufacturer of heavy earth moving equipment might want to be able to monitor its equipment to understand its long term health and performance; or go beyond and manage their warranty, contract enforcement, cross sell/upsell etc. IoT can enable this and more.
What’s more, connected devices don’t have to be particularly complex to enable actionable insight that can dramatically change a business’s fortunes. The journey can start small and then scale up driven by needs and benefits. For example, data from simple, low-cost retrofitted devices in a retail store or shopping malls can help them monitor and manage energy costs. Similarly, data about footfall and customer movements can help improve target promotions for customers, increasing revenue and also profitability.
The IoT undoubtedly represents a huge opportunity for businesses right across the spectrum. While many have largely bought into the idea, most aren’t yet able to apply sophisticated levels of predictive and prescriptive analysis to the data that the connected devices are producing. To truly maximise the potential of the IoT and achieve business outcomes like greater productivity and efficiency, and lower costs or to increase revenue, companies must strive to apply more advanced data analysis.