Boosting customer efficiency with IoT, predictive analytics and machine learning

Hitachi Consulting is the strategic business consulting division of Hitachi, Ltd. It has offices in North America, Europe, the Middle East and Asia – with global delivery centers in India and China. “We use IoT, predictive analytics, and machine learning to make the operations of our customers more efficient, says Feroze Mohammed, executive vice president, Hitachi Consulting Software Services India. In conversation with Ankush Kumar, Mohammed shares global case-points of tech implementation to address most complex issues and how the IT solution providers are shifting towards outcome based model for charging customer.

Tell us about the key focus areas of the company? What are the major technologies that you are using ?

We are trying to build solutions in four different categories; digital operations, digital customer experience, digital enterprise, and smart infrastructure. Digital operations is about digitizing the operations of manufacturing and supply chain. This is how we work on making smart factories and trying to build predictive maintenance of equipments. We use IoT, predictive analytics, and machine learning to make the operations of our customers more efficient.

The other category is digital customer experience; which is more about how we enhance customer experience using digital technologies. The third is the digital enterprise that is more about implementing cloud and automation. The fourth portfolio is called the smart infrastructure which is about building OT/IT and IoT infrastructure. This is a more focused approach towards public safety, good governance, smart cities, smart water management, smart energy management, etc.

How much do think is the scope in big data as a business opportunity ?

I think big data is not being used properly and there is no mechanism to actually reach to this data and make sense out of it. Most of the organizations and government bodies collect phenomenal amount of data at the grass root level. Can we access this data, can we make decisions in the real-time based on that data. Lets take a hypothetical scenario, lets say we have soil data, rock data, rock patterns, water data of the state, weather data with future predictions – so can we tell our farmers that for this season you will be better off going in for this crop in this region because this is the water level, this is the soil condition, this is the kind of weather that is forecasted for the next year and you are better off doing this kind of produce. That will certainly be a huge advantage for the farmers.

Can you give us case points where you have emphasized on making the data more valuable?

We are doing some work for Copenhagen city, Denmark. What we are doing there is building a city data exchange market place as a city generates a lot of data. For example a city like Mumbai generates lots of data on how the trains are running, timing of the trains, timing of the buses, etc. and lots of different aspects of the city like pollution, energy consumption, water consumption by locality release data points. So what we are doing is we are creating a city data exchange. The idea is that you create a data exchange where all the city data is available in the market place for other solution providers to use this data. We are also working on a UK rail project which is part of the smart infrastructure services that we provide. UK rail is building about 122 state of the art intelligent trains for their UK intercity program and we have been asked to run and maintain this for 20 years plus. For us, its not just about providing these locomotives etc, we will have to figure out a way of how the uptime of the network is going to be built. The higher the uptime of the network the better will be the gains.

What kind of solutions you have for the smart city project ?

We have an interesting software called Hitachi Visualization. It has got an ability to analyse the video footage coming from the CCTV, like traffic videos. This is used with the combination of sensor data, it applies facial recognition and object recognition. We are working on this technology that can help government in a very big way in future, so some of the obvious areas are security and safety, but analytics is also going to help in a big way.

Do you think legacy infrastructure is still a challenge for large organisation?

That’s what we feel is going to be a unique problem for India. In India people are not going to throw away their investments on existing machinery, So these are called black machines. Black machines don’t have the ability to collect data, so how can we actually build a sensor on the top where the existing infrastructure is protected. We collect data and make an analysis based on that data which is generated from sensors embedded on top of the black machine. We would like to focus on both in terms of video analytics of the manufacturing plants and IoT sensors on this existing infrastructure.

Do you see a change in business model in the IT industry? What business model have you adopted ?

We believe that the future world is going to be more about offering solutions to propel business rather than just solving a problem at hand. Now customers are moving from complete end-to-end solution model to an outcome based model. So in this model we don’t just have to build the overall solution but to also run and maintain the solution. For us to monetize it, the payback period is much longer and we have to run the solutions effectively to make money on that.

We are working on outcome based model. There is lot of solutions that has already been sold across the world in terms of outcome based model. Like in AT&T, we manage all their power consumption of data centers in US. We baseline the energy bill. And then we will take it over and will be responsible for managing energy and whatever savings happen we take a split between both of us. And we essentially replace all the kind of gear, switches that are required for the electrical equipments including small little things like bulbs. We are able to do a lot of analytics and power management in a smart way so that the total energy consumption comes down. And we have to run and maintain it for the next ten years. So whatever we gain on this gets split between the customer and us. Models like this is only suitable for large sized organization as it requires the ability to wait for five or ten years till you get your returns.

 

Hitachi ConsultingIOTmachine-learningpredictive analytics
Comments (0)
Add Comment