Geospatial Predictive Modelling for the Armed & Security Forces

Geospatial Predictive Modelling for the Armed & Security Forces

SHARE

Security related incidents always have a spatial component associated with it and with GIS technologies – Where, What, When, Why, Who and How are being analysed continuously. The past incidents and the happenings around us clearly lead to one conclusion that time has come when our approach needs a paradigm shift from reactive to proactive.

Continuous observation of the AOR is an absolute requirement for the success of any kind of ops from conventional to CI/CT to NCW. For sound decision-making and advancing own plans to achieve success in the planned mission mil leaders and security forces are becoming more and more dependable on technology because of the use of technology by the ANEs against whom the mil leaders and security forces plan their ops. An innovative way to achieve better situational awareness could be by fusion of Big Data Applications and GIS technologies. Intelligence collected today is from diverse sources and are in different formats. To analyse this huge inputs that too in different format in a manual way is not only a tedious process but also cannot be delivered to the commanders at all levels in the time and place they want. Decision making which is a critical function of command suffers.

Predictive analytics is an area of data mining that deals with extracting information from data and using it to predict trends and behaviour patterns. Often the unknown event of interest is in the future, but predictive analytics can be applied to any type of unknown whether it be in the past, present or future. The core of predictive analytics relies on capturing relationships between explanatory variables and the predicted variables from past occurrences, and exploiting them to predict the unknown outcome.

Geospatial Predictive Modelling is rooted in the principle that the occurrences of events being modelled are limited in distribution. Occurrences of events are neither uniform nor random in distribution – there are spatial environment factors (infrastructure, sociocultural, topographic, etc.) that constrain and influence where the locations of events occur. Geospatial predictive modelling attempts to describe those constraints and influences by spatially correlating occurrences of historical geospatial locations with environmental factors that represent those constraints and influences. Geospatial predictive modelling is a process for analysing events through a geographic filter in order to make statements of likelihood for event occurrence or emergence.

There are two broad types of geospatial predictive models – Deductive and Inductive.

► Deductive method — The deductive method relies on qualitative data or a subject matter expert (SME) to describe the relationship between event occurrences and factors that describe the environment. As a result, the deductive process generally will rely on more subjective information. This means that the modeller could potentially be limiting the model by only inputting a number of factors that the human brain can comprehend.

An example of a deductive model is as follows. Sets of events are typically found :-

  • Between 2000 and 3000 meters from the LOC.
  • In the dense forest land cover category.
  • At elevations between 1100 and1600 meters.

In this deductive model, high suitability locations for the set of events are constrained and influenced by non-empirically calculated spatial ranges for LOC, land cover of dense forest,and height. Lower suitability areas would be everywhere else. The accuracy and detail of the deductive model is limited by the depth of qualitative data inputs to the model.

► Inductive method —The inductive method relies on the empirically-calculated spatial relationship between historical or known event occurrence locations and factors that make up the environment (infrastructure, socio- culture, topographic, etc.). Each event occurrence is plotted in geographic space and a quantitative relationship is defined between the event occurrence and the factors that make up the environment. The advantage of this method is that software can be developed to empirically discover – harnessing the speed of computers, which is crucial when hundreds of factors are involved – both known and unknown correlations between factors and events. Those quantitative relationship values are then processed by a statistical function to find spatial patterns that define high and low suitability areas for event occurrence.

Inductive method will suit the armed and security forces better since it leverages statistics to predict outcomes. Most often the event one wants to predict is in the future, but predictive modelling can be applied to any type of unknown event, regardless of when it occurred. For example, predictive models are often used to detect crimes and identify suspects, after the crime has taken place.

Smart Map Grids for Geospatial Predictive Modelling

GIS Maps for geospatial predictive modelling will have the thematic layers of the sensors deployed in the target area. Each thematic layer will have the spatial information in terms of its coverage be it line of sight, range analysis, fan analysis etc. on land, water and space. These maps should have the flexibility to be stitched to encompass the whole area of the country and the aim plus would be to have this data about the trans-border areas too where the actual war is going to be fought. Physical potential of the environment in terms of bio-climate, landforms, geology as well as vegetation will also form part of thematic layers as explained earlier. These maps are going to be more detailed in whatever resolutions it exist.

Another feature of this map will be it will undergo dynamic changes as per the change in the pattern of deployment of the sensors. These could be done by each country and then at a global forum the modalities for exchange of such data to fight militancy can be worked out leading to eradication of militancy, thus, making our planet a better place for all human beings to co-exist. The climate data from all weather stations located at various parts of the world should also form part of this data set. Voids if any in the absence of weather stations can be filled with interpolation techniques. Actions on this line will lead

to standardisation, which is a must to fight terror on a global platform since it is easier to understand, if, all speak the same language. Since this data set is going to be huge it will require Big Data applications to gain insight and that is where predictive analytics will play a significant role.

Data fusion is the process of integration of multiple data and knowledge representing the same real-world object into a consistent, accurate, and useful representation. The expectation is that fused data is more informative and synthetic than the original inputs. Sensor fusion is also known as data fusion and is a subset of information fusion. In applications outside the geospatial domain, differences in the usage of the terms, data integration and data fusion apply. In areas such as business intelligence, for example, data integration is used to describe the combining of data, whereas data fusion is integration followed by reduction or replacement. Data integration might be viewed as set combination wherein the larger set is retained, whereas fusion is a set reduction technique with improved confidence.

Benefits of Geospatial Predictive Modelling

A sensor deployment map doesn’t exist as of now. On such a thematic map, application of geospatial predictive algorithm will give out the voids existing if any in spite of the sensor deployments including observation posts giving out bio–feeds. This void could be routes of infiltration which can be secured immediately by re-alignment of sensors. If this concept is applied on the satellite imagery of trans-border areas and if the feeds from Drones and Satellites Images are captured and

analysed for frame by frame changes we can peep through the enemy activities.

Following additional benefits in terms of battlefield transparency can be achieved:-

► Qualitative improvement in operations. The combination of asset visibility along with big-picture analytics and insights will lead to better situational awareness and enable commanders to take better decisions. Live digital model of area that is always accessible, always fresh, and completely fused together in a common operating picture will provide an avenue for application of spatial predictive analytics that learns from experience (data) to predict the future behaviour of individuals in order to drive better decisions.

► Advanced Failure Detection. When we monitor the critical devices, machines, vehicles etc. with RFID tagged sensors and apply predictive analytics on the pattern of fault generated we can get real-time maintenance alerts that enable us to avoid unexpected failures of devices, equipment, and systems. Identifying unseen problems and predicting failures before they happen will change our maintenance philosophy. When we are able to utilise predictive maintenance, we save the time and money required for unnecessary maintenance.

► CI/CT Operations — Big data collected by drones, satellites, UAVs, GIS based Applications and technical intercepts etc as part of intelligence, surveillance and reconnaissance (ISR) data can be analysed automatically using predictive modelling. This will allow the CI/CT operations to be carried out in near real time.

► Cyber Security — Big Data analytics can be applied to spot advanced persistent threats – such as socially engineered attacks designed to steal government information which has happened in the past (Chinese attack on our websites). Most hackers have a modus operandi, which once identified can be used to predict the form of future attacks and put appropriate defensive measures in place.

Geospatial Predictive Modelling Use Cases in the Commercial Sector

Following use case merits attention:-

► Location for Establishing Malls. The demographic advantage can also help companies that have a long-standing reputation among their existing clientele. By tweaking their marketing strategy to target a new audience they can potentially refresh their brand, keep it modern, and shed light on an aspect of their brand that might not have been as appealing to existing customers.

► Enhancing Public Health— From the common cold to a life-threatening emergency, access to healthcare is crucial to ensuring a healthy society. The marriage of Big Data and GIS is bringing healthcare into the future, with many benefits to the public. Businesses and local government alike can use GIS to determine what services need to be rendered based on local health trends, and locate new sites for healthcare facilities. The public benefits from this by having increased awareness and access to nearby facilities ready to address their healthcare needs.

► Creating Emergency Action Plans — Big or small, emergencies affect us all at some point. Public service cornerstones are tapping into the powers of predictive analytics to create action plans in the event of an emergency. Knowing when to head off a potentially dangerous situation or mitigate damage is extremely useful. From military bases to fire departments to pharmacies and hospitals, predictive analytics can help map out the disaster management plans.

An Implementation Strategy

Implementation should follow an evolutionary path within the wider Information Management/Information Exploitation Strategy and be coherent with the Defence ICT Strategy. One should not rely solely on bespoke geospatial predictive modelling capabilities for defence forces. Rather, the defence forces should build on the huge investment in this area being made by the commercial sector and in doing so ensure that the defence forces is well-positioned to track and exploit further commercial technological developments as and when they occur. It can be started small using existing data with which the defence forces are comfortable. Things should be kept simple initially and then migrate to more complex uses subsequently. Focus on day to day functional issues since this will help build a use case for the usefulness of geospatial predictive modelling.

Conclusion

Network centric warfare, manoeuvres warfare, counter terror operations, counter-insurgency operations and covert operations all stand to gain from geospatial predictive modelling since it offers options and permutation/ combinations to decision makers to firm on plans and strategies. Big data can combine multiple data sources in the same view with dashboard like applications in real time. It can show where of the data through smart mapping. Scope for predictive analytics and forecasting can overcome operational challenges in real time. Ground, air, sea and strategic forces will be assisted in conduct of their operations. Geospatial Predictive Analytics platform will analyse the patterns and can be a valuable input for operations to be carried out. It can be a Web based application and will be able to generate report which will be shared across the stake holders.