Evaluating workplace well-being
Employee wellbeing and productivity of the organisation are closely interlinked concepts mutually impacting each other in the long run. The idea of workspace wellness and health promotion has been making rounds in recent times, and the outset of the coronavirus pandemic has given added importance to the topic. Having conducted intense study in this regard, Prof AVS Kamesh and his PhD scholar Ms Ashrafunnisa Mohammed from the Department of Management have authored a chapter titled ‘’Evaluating the Role of Workplace Well-Being in Enhancing Employee Productivity in the 21st Century’’, in book Transforming Lives through Mindfulness published by IIM Bodh Gaya AND Excel India Publishers.
Their study tries to evaluate the unique role of workplace well-being in enhancing employee productivity in the 21st century. The contemporary business world is characterised by organisations competing against each other. In a dynamic and complex business environment, the main aim of modern corporate organisations is to retain talented employees by ensuring mindfulness and workplace well-being because these employees are the main sources of competitive advantage from a strategic point of view. Thus, an enhanced level of workplace wellbeing is an effort to create happy and productive workers so that they work optimally and happily. Work is a key social determinant of population health and well-being.
Workplace well-being in India is often focused on changing individual health behaviours through employer wellness programs. The Covid-19 health crisis brought into focus, some of the limitations of present approaches revealing structural conditions that intensify the physical and psychosocial problems of employees and their family members. The onset of the Covid-19 pandemic augured the new dimension of workplace well-being by converting home into a happy place to work. This perspective leads management experts and human resource managers to think about a combined model of work and home as a place of well-being.
This chapter is significant in the times of Post Covid-19 with new perspectives evolving to discuss work-life balance and well-being at workplace which are significantly related to the productivity of the employees. The article mainly targets management educators, corporate managers and personnel managers from public sector companies in India, calling for a comprehensive renovation in the workplace environment.
- Published in Departmental News, News, Paari Current Happenings, Research News
Co-edited and authored in the world’s longest-running science journal
Dr Soumyajyoti Biswas, Assistant Professor, Department of Physics, had two lucky breaks as he got his article “Kinetic Exchange Models of Societies and Economies” featured in the prestigious journal Philosophical Transactions of the Royal Society A, the theme issue co-edited by Dr Biswas himself, along with Dr Guiseppe Toscani from the University of Pavia, and Dr Parongama Sen from the University of Calcutta. Philosophical Transactions of the Royal Society has the prestige of being the world’s longest-running science journal launched in 1665. Publishing high-quality theme issues on topics of current importance and general interest within the physical, mathematical, and engineering sciences, the journal continues its history of influential scientific publishing.
A kinetic model of binary interaction, with conserving or non-conserving exchange, has been an elegant and powerful tool to explain collective phenomena in myriad human interaction-based problems, where an energy consideration for dynamics is generally inaccessible. Nonetheless, in this age of Big Data, seeking empirical regularities emerging out of collective responses is a prominent and essential approach, much like the empirical thermodynamic principles preceding quantitative foundations of statistical mechanics.
Through this theme issue, the authors intend to bring together the current progress in the applications of kinetic exchange models in various applications of societies (opinion formations, rating, social networks, fake news, etc.) and economies (inequality measures, taxation, trade models, behavioral economics, etc.) using numerical simulations, machine learning techniques, analytical methods, and data analysis, reported by physicists, social scientists, mathematicians and economists through some of the original and reviewed articles.
In human interactions, such as a trade (exchange of money) or, discussions or debates (exchange of opinions), following simple dynamical rules, a collection of agents (a society) shows emergent properties that are widely seen in real data (distributions of wealth, formation of consensus, etc.). Without knowing the complexities that are involved at the individual levels, it is, therefore, possible to understand the average properties of the society as a whole. This is reminiscent of simple elastic collisions of ideal gas molecules that give average thermodynamic properties, such as temperature, pressure, etc. without knowing the complexities of the individual atoms. This has been a widely followed route to formulate statistical physical models of societies and economies.
The kinetic exchange models have been a very successful set of tools to understand the socio-economic emergent properties from simple models. Among other things, these models helped understand the growth of economic inequalities, the effects of taxes as well as the spread of opinions. A close quantitative resemblance with real data from various countries of the world demonstrates its usefulness.
The future prospects of the kinetic exchange models for societies and economies include possible predictions of extreme fluctuations in average measurable quantities by looking at the inequality of time series data. The models can help us in identifying the features of the real data that can mirror the underlying extreme fluctuations.
- Published in Departmental News, News, Physics News, Research News
Department of ECE publish paper on Fourier holography
A paper titled “Sparse reconstruction for integral Fourier holography using Dictionary Learning method” has been published by Dr Inbarasan Muniraj, Dr Karthikeyan Elumalai and Dr Sunil Chinnadurai – Assistant Professors of Electronics and Communications Engineering at SRM University-AP, along with PhD students Lakshmi Kuruguntla and Vineela Chandra Dodda.
The paper proposes reconstructing holograms from fewer data, thereby reducing the need for processing the complete hologram data, which is otherwise computationally expensive.
Abstract: A simplified method was demonstrated to generate a hologram from multiple two-dimensional (2D) images. Sparse reconstruction was shown using the Sequential Generalised K-means (SGK) algorithm. It is shown that the proposed sparse reconstruction method provides a good hologram quality, in terms of peak signal-to-noise ratio, even under ~90% sparsity.
The paper is written in collaboration with Professor John T Sheridan, Vice-Principal for Research & Innovation – College of Engineering & Architecture, Head of School of Electrical and Electronic Engineering, University College Dublin, Ireland.
Holography has been shown useful for biomedical imaging, cryptography, data storage, and entertainment. The future plans of the research group include extending this approach to other holographic systems such as digital holography and holographic microscopy.
- Published in Departmental News, ECE NEWS, News, Research News
Network resource allocation for emergency management
The focus of the network keeps changing with every generation of communication technology. The 5G era is waiting for the next generation to bring a remarkable revolution in communication technologies. Applying various changes and modifying the drawbacks of 5G technology can help us to improve the features of the upcoming 6G. The Department of Electronics and Communication Engineering is delighted to inform you that the paper, “Network Resource Allocation for Emergency Management based on Closed Loop Analysis” has been published by Dr Udaya Sankar, Assistant Professor, and BTech students; Sai Jnaneswar J and VMVS Aditya, in “ ITU Journal on Future and Evolving Technologies – 2nd special issue on AI and machine learning solutions in 5G and future networks”.
Abstract of the research
Telecommunication systems being a critical pillar of emergency management, intelligent deployment, and management of slices in an affected area, will help emergency responders. Techniques such as automated management of ML (machine learning) pipelines across the edge and emergency responder devices, usage of hierarchical closed-loops, and offloading inference tasks closer to the edge, can reduce latencies for first responders in case of emergencies. This paper describes the major findings of building a Proof of Concept (PoC) for network resource allocation for emergency management using a hierarchical autonomous artificial intelligence (AI)/ML-based closed-loops in the mobile network, which was organized by the Internal Telecommunication Union Focus Group on Autonomous Network (ITU FGAN). The background scenario for this PoC included the interaction between a higher closed-loop in the Operations Support System (OSS) and a lower closed-loop in RAN (Radio Access Network) to intelligently share RAN resources between the public and emergency responder slice. Representation of closed-loop “controllers” in a declarative fashion (Intent), triggering “imperative actions” in the “underlay” based on the intent, setup of a data pipeline between various components, and methods of “influencing” lower layer loops using specific logic/models, were some of the important aspects investigated by various teams. The main conclusions are summarised in this work, along with the significant observations and limitations from the PoC as well as future directions.
Explanation of the research
This is a collaborative study where the researchers have developed and implemented a hierarchical closed-loop that autonomously handles an emergency case. This project contains several groups working on separate functions such as monitoring, computing, ML selection, and resource allocation. Some presented how the AI agents and the network can be adapted to assist mobile network users in Search, Diagnostic, and Rescue (SDAR) missions. Some integrated the implementation of the closed loops in O-RAN based software. The researcher’s role in this project is to generate the data which will be used for analysis and this gets integrated with another team. Rather than putting dummy data and analysing or creating and training a machine learning model, we use the Simu5g simulator which is library-based on OMNET++ framework to generate data, mainly resource allocation data. In Simu5g they configure the required network, set all the input parameters such as and simulate it for a specific duration of time. After simulation, the various results are obtained like SINR, throughput, resource block allocation, and many more. This data of the selected parameter results is converted into CSV files and handed over to the required for ML. This project blends all the chunks of work like monitoring, computing, ML selection, and resource combination, to facilitate network resource allocation for emergency management.
Practical implementations of the research
In this project, the researchers have gathered the simulated data and want to analyse it and apply machine learning algorithms to improve the condition of existing 5G networks. AI/ML model will help to analyse the data which helps in optimizing resource allocation. Optimization algorithms for resource allocation like Interference minimization, Throughput maximization, and many more. Optimising such parameters would be really helpful for the development of new generation networks.
Collaborations
The paper was developed as part of a build-a-thon track, “ITU AI/ML in 5G Challenge: applying machine learning in communication networks” challenge. The researchers worked under the problem statement “ITU-ML5G-PS-014: Build-a-thon (PoC) Network resource allocation for emergency management based on closed-loop analysis”. This project is guided by Dr. V Udaya Sankar under AI-Designed Wireless (AIDW) Project Simulated network scenarios using simu5g for measuring resource allocation. This build-a-thon challenge was started on June 7, 2021 and ended on November 5, 2021. The research team was placed fifth among 12 teams who participated across the world in the “build a thon” track of the global challenge. While the overall challenge saw over 600 participants across more than 80 countries, the “build a thon” track was a unique coding challenge from ITU FG AN. The team worked on the simulation of the 5G system and corresponding analysis using Simu5G from ITU global partners. After the build-a-thon, the teams that have participated in the build-a-thon under the problem statement “ITU-ML5G-PS-014: Build-a-thon (PoC) Network resource allocation for emergency management based on closed-loop analysis” were collaboratively developed the journal from the observations and results that were made during this build-a-thon.
The current research work focuses on the Implementation of a simulation environment to generate data for model training and testing purposes, and serve as a simulation underlay for testing. The simulations allow us to study various configurations and analyse them to optimize the allocations. For the later stages, various machine learning techniques can be applied for the data that was generated through simulation, and further, the comparison can be done for various machine learning techniques.
- Published in Departmental News, ECE NEWS, News, Research News