A lot of innovation is coming from start-ups who are not primarily building geospatial products, but are actually the users of geospatial in what they are trying to do, Sandeep Singhal, Director, Cloud Storage, Google, USA, highlights in an exclusive interview
GIS has long been playing a vital role in Business Intelligence. What are your views on this?
The importance of GIS can be stressed enough to provide valuable insights into the business, identify potential areas that are worthy of an analysis and then help present those results effectively back to customers. It is both a pre-analysis layer to understand what is happening and then a presentation layer to presenting it back to users. A lot of heavy lifting work is happening in the areas of analytics and machine learning using data that is pulled from the GIS systems and then delivered right back to the GIS systems.
Geospatial technologies are making Business Intelligence more intelligent, enabling businesses to take better decisions. What do you think the trends in this regard are?
The trend here is really towards deep learning and deep understanding of information, getting a large variety of information from different sources, ranging from small sensors to traditional mapping applications, and delivering that knowledge to enable you to make decision and make changes. Business Intelligence and the Fourth Industrial Revolution go hand-in-hand. There are tons of examples where companies are using images and image process to extract information from satellite imageries and drone imageries to understand various aspects like crop behavior — where does fertilization need to happen; what is really happening in the fields. It is important to bring in machine learning to understand what is happening to these images, add real-time imagery from drones and other local sources and then drive decisions that a farmer can take to optimize his crop yield. So, this is a form of Business Intelligence that takes advantage of all of these different components.
With the expansion of the geospatial industry, there is more and more inclusion of building tools for analyzing geospatial data. Many are of the view that the core geospatial industry is being nibbled on two sides by IT and large industrial powerhouses, with both segments developing their own geospatial capabilities. However, I feel that this is a great opportunity to partner more effectively to deliver what customers really want, very effective and real-time services.
A lot of innovation is coming from start-ups who are not primarily building geospatial products, but are actually the users of geospatial in what they are trying to do. They are embedding data, maps, etc. as part of their core applications. What they are looking for is ways to build their apps more quickly, and more efficiently, and build better experiences by using geospatial technology. A key factor here is location intelligence. It is important to reach out to them with an objective to make geospatial data easy to access, quicker to access, and more effective in the applications that these companies are trying to build.
How technological advances are facilitating use of spatial analytics in business intelligence?
Three factors are important here. One is the ability to capture data very quickly from lots of different sources. We are seeing a whole generation of small light-weight sensors coupled with communication capabilities that can move data in and out of everything, like from a piece of equipment to sensors embedded in ground or inside buildings.
Second is the effective infinite storage and computation power provided by the Cloud. The ability to take data from anywhere, aggregating it, and doing as much analysis you possibly want to do using Cloud provides computational power. There has been an exponential surge in Cloud providers who provide core infrastructure for supporting geospatial services.
The third key element is the growth of very rich analytics, Big Data and machine learning libraries that really democratize how you analyze data. The technology makes it possible to very quickly develop machine learning models based on the large volume of data that is being fused in the Cloud and draw conclusions that can be actionable. This combination of cheap sensors, ubiquitous sensors, massive amount of storage in computation and standardized analytics and machine learning are really powering this revolution.
Geospatial visualization and analytics have already made roads into a wider swath of industries. Do you still see some considerable challenges?
Geospatial visualization and analytics is becoming ubiquitous. We are seeing more and more users visualizing geospatial data either simply using standard off-the-shelf maps like Google Maps Engine, or Google Earth to visualize information, or using commercial tools like ArcGIS. I think the trend here is towards making more geospatial data available to mobile devices whether using commercial or off-the-shelf consumer tools or commercial tools. The consumer geospatial data is no longer confined to a PC or workstation. They want their data now, they want in their hand, they want it on a device which they are walking around with. So, the delivery is really shifting to very rich geospatial data in a mobile form factor in real time when and where people need it.
Are security and regulatory factor still barriers for Cloud implementation across verticals like government and financial services?
There are still regulatory requirements for any Cloud implementation. Some are industry specific like in the financial industry there is a number of regulations about data protection, data archival and data recording. Governments also have various requirements regarding privacy and security. So, regulations do exists. Most Cloud providers including Google, Amazon and Microsoft are working aggressively to make sure that their Clouds do support all of the active regulatory requirements both at industry level and at geographic level. So, the gaps are getting closed to rapidly. we are also beginning to see that almost every industry is recognizing that those gaps are no longer unsurpassable and they can actually begin to move critical workloads into the Cloud.
We are on the verge of the Fourth Industrial Revolution. What role will geospatial technologies play in this revolution?
The Fourth Industrial Revolution is the marriage of data, analytics and real-time presentation to allow customers to make decisions and improve the efficiency of their processes. The key enablement for 4IR is the fact that we are now bringing together massive amount of data, applying deep analytics, Big Data analytics and machine learning to fuse the data together and draw conclusion and then present that information very quickly in order to drive changes in how we manufacture, schedule resources and so on. This is a great fundamental shift.
A great example is a company called Cartogram, which is working in the medical sector, helping hospitals understand where their resources are — be it human resources or medical equipment such as X-rays machines etc. Using data about location and what procedures are necessary, an analytics can be carried out to understand the geometry of the venue and then a real-time guidance on how to move patients to the closest medical equipment can be presented. This helps in geting the right procedures done as quickly as possible. This is a great example of how data and analytics and presentation deliver a totally new way of delivering services.