Health Care:

With data science and optimization models, public-health system could be transformed into a high-quality and quick-response service-delivery system. The data analytics, AI and optimization-based solutions have great potential in bringing efficiency and effectiveness into service-delivery. The approach could include analysing the past information at primary health centre (PHC) level.  With the help of machine learning models, a predictive health-care system could be developed. Further, simulation and optimization models would help is the planning and operations management of primary health-care service supply chains. For example, a possible web-based TB registration scheme could generate massive data by enhancing the outreach of the health services. Network of telemedicine services, mother-and-child tracking system of weekly voice-messages to pregnant women and neo-natal mothers could be facilitated. With more than one billion mobile users in the country, mobile phones offer tremendous opportunities in the efficient health-service delivery, which could include:

• At any individual level, a predictive approach to provide proactive primary health services to avoid severe health complications;
• Ensuring the patient-tracking and treatment-adherence by sending messages to patients with diabetes or HIV/AIDS reminding them to take treatment;
• Quick-reporting of cases during outbreaks or epidemics;
• Alerting next-level of health services regarding emergency situations such as difficult or complicated labour cases in remote rural areas;
• Advanced predictive-response with optimal ambulance routing system could reduce the time of reaching for emergency health services.

Publications:

 - A Generalized Epidemiological Model for COVID-19 with Dynamic and Asymptomatic Population, arXiv:2011.09686 [q-bio.PE], 2020 (A Ghatak, SS Patel, S Roy, and S Bonnerjee)

- Application of ML Algorithm for Controllling MMR of Andhra Pradesh (SS Patel and Ankit Deshia) (Working Paper)

Natural Resource Management (water, food &energy) :

This IDeAL will enable researchers to apply decision science and data science models against real-world challenges such as water storage, biodiversity loss and the extraction of mineral resources. And the level of cross-disciplinary collaboration to address the grand challenges of water, food, and energy security for increasing population as a grand challenge.

The management of all-natural resources faces the common central problem of how data is exploited to build predictive and integrated models that can be used to make sustainable decisions in the presence of uncertainty.

Advanced statistics, math, coding can reveal complex, interdependent relationships between global water-resource features, poverty, and energy consumption rates. Rainfall variability, droughts, massive flooding it is related to a lack of sustainable water resources for agricultural development, more runoff and erosion, and overall decreases in that nation's GDP. So, using Natural resource data science, able to draw strong correlations between a nation's rainfall trends and its poverty rates.

To provide techniques in integrated modelling frameworks with novel data science techniques and, in particular, innovative combinations that can make sense of the increasing complexity, variety, and veracity of underlying Natural resource data, also exploiting multiple data sets including real-time streaming data—at the same time, incorporating a sophisticated spatial and temporal reasoning across scales, as an integral aspect of natural resource data science and not something that just provided through separate tools such as GIS tools.

Publications:

- “A Bilateral River Bargaining Problem with Negative Externality”, arXiv:1912.05844, 2019 ( SS Patel and Parthasarathy Ramachandran)

- “Comparison of Machine Learning Techniques for Modeling River flow time series: The case of upper Cauvery river basin”. Water Resour Manage 29, 589–602 (2015). https://doi.org/10.1007/s11269-014-0705-0 (SS Patel and Parthasarathy Ramachandran)

Emergency Management:

The advanced data analytics solutions will be able to cope with this global issue by exploiting heterogeneous data sources. People and automatic systems generate massive data during natural hazard events (e.g., social network data generated by citizens and first responders, satellite images of the affected areas, flood maps generated by drones). This heterogeneous information can be transformed into valuable knowledge by integrating it and extracting knowledge in near-real time through data analytics solutions.  Research at IDeAL will bring together state-of-the-art theories in data science, artificial intelligence, machine learning, open-source data, and crowd-sourcing improve decision-making during disasters.

• Hazard Nowcast and Forecast models based on the integration of earth observations, social media, and crowd-sourcing
• Climate Change Models, Risk Maps for supporting emergency management
• Pandemeic Management

Publications:

 - A Generalized Epidemiological Model for COVID-19 with Dynamic and Asymptomatic Population, arXiv:2011.09686 [q-bio.PE], 2020 (A Ghatak, SS Patel, S Roy, and S Bonnerjee)

Computational Public Policy:

Alice Rivlin mentioned this asepect in 1970, when she published “Systemic Thinking for Social Action.” Argued for more rigor and scientific processes in government decision-making, Rivlin wrote a pithy final line: “Put more simply, to do better, we must have a way of distinguishing better from worse.”

The data science approach is tremendously valuable for public servants and public policy. It pushes people to defy conjecture, consider counterfactuals, reason about complex patterns, and question what an information is missing. It makes people skeptical of tales, which, while often emotionally powerful, are not good sources of information on which to build comprehensive policies. Computational public policy is the application of computer science or mathematics to solving problems in public policy.The computational public policy includes, but is not limited to, principles and methods for public policy formulation, decision making, analysis, modeling, optimization, forecasting, and simulation.

Image Source: Alex Engler/The University of Chicago

Publication:

“Optimization and Policy Analysis of Water allocation in a River Basin”, Sustain. Water Resour. Manag. 4, 433–446 (2018). https://doi.org/10.1007/s40899-017-0124-5 (SS Patel and Parhasarathy Ramachandran)

Cyber-physical systems:

As computing and communication devices become smaller and cheaper, they can be embedded in objects and structures to interact directly with the physical environment and reach human capabilities. Cyber-Physical Systems (CPS) are a compilation of computing and communication units that connect the cyberworld of computing and communications with the environment. CPS spans applications with enormous societal impact and economic benefit. Cyber-Physical Systems are characterized by many tightly integrated complex components in a network, expanding and contracting dynamically. CPS is already used in medical devices (pacemakers, insulin pumps), infrastructure (surveillance and control), manufacturing, transportation (airplanes and air-traffic control, rail). Advances in CPS will make it probable to build systems that will significantly surpass the capabilities of the simple embedded systems of today.

The IDeAL is currently focusing research in the area of:

• energy management
• network security
• smart systems
• system resource allocation
• Infrastructure

Publication: 

“Forecasting health of complex IT systems using system log data”, J BANK FINANC TECHNOL 4, 27–35 (2020). https://doi.org/10.1007/s42786-019-00011-z.