AI for Decision Intelligence in Critical Systems (Advances in Applied Mathematics)

This book is a multi-disciplinary reference on how domain-aware artificial intelligence (AI) models can outperform generic approaches by addressing sector-specific complexities. It offers comparative frameworks, reproducible case studies, and real-world applications of emerging AI methods.

Collectively, AI for Decision Intelligence in Critical Systems emphasizes a unifying theme: the effective deployment of AI to strengthen decision-making, enhance system reliability, and mitigate risks in domains where precision, trust, and efficiency are critical.

This edited volume brings together twenty-one chapters of original research, each exploring how AI, machine learning (ML), and deep learning (DL) are shaping innovation across critical domains. The book highlights the application of advanced architectures―including Convolutional Neural Networks (CNNs), Quaternion Neural Networks (QCNNs), Large Language Models (LLMs), and Gradient-Boosted Decision Trees (GBDTs)―to solve complex, domain-specific challenges.

Concerning computer vision and infrastructure safety, chapters discuss the use of CNNs and QCNNs for automated road crack detection, offering scalable approaches to improving transportation safety while reducing dependence on manual inspections. With regard to software engineering, contributions focus on leveraging ML, DL, and LLMs to enhance software quality assurance, minimize defects, and improve resilience in high-stakes industries. Additional chapters examine ML-driven methods, particularly GBDT, to uncover non-linear drivers of equity valuation across sectors, supporting more accurate forecasts and risk-sensitive decision-making.

Academics and researchers in computer science, AI, and data science, industry professionals in transportation, software engineering, finance, and policymakers seeking to apply AI systems effectively will find this book useful.

Dr. Shahab Saquib Sohail is an Assistant Professor in the Department of Computer Science and Engineering at Jamia Hamdard, New Delhi. He previously served as a Senior Assistant Professor at VIT Bhopal University. He holds a Ph.D. in Computer Science from Aligarh Muslim University. He is recognized among the top 5 researchers globally in the Scopus database for work related to ChatGPT and among the top 2% of AI and Computer Vision researchers worldwide (Stanford–Elsevier list). He has authored more than 100 SCI-indexed journal papers, including 70 Q1 and Q2 publications. His research has appeared in high-impact venues such as Nature Machine Intelligence, Information Fusion, IEEE Transactions on Big Data, and WIREs Data Mining and Knowledge Discovery, as well as leading conferences including INTERSPEECH, IJCNN, ICASSP, and ICDM workshops. With over 3,000 Google Scholar citations, his research spans computational intelligence, recommender systems, and computational social science. Dr. Sohail is also an active collaborator with international research groups and a dedicated mentor to emerging scholars in AI and machine learning.

Arpita Soni is a senior IT professional with over two decades of experience in software engineering, quality assurance, and program management. She specializes in generative AI, automation, and digital transformation across banking, healthcare, and supply chain sectors. A certified Project Management Professional (PMP) and Certified Scrum Master (CSM), she is also a Senior Member of IEEE and a Fellow of the British Computer Society (BCS). Arpita has led large-scale enterprise AI initiatives that enhance operational efficiency, compliance, and reliability. She has authored multiple research papers on AI and machine learning, including work on low-resource chatbots and AI integration into software development lifecycles. She is a frequent keynote speaker, session chair, and reviewer for leading IEEE, Elsevier, IGI Global, and Springer journals and conferences. Arpita is also the author of books such as AI Sustainability and Advanced Statistical Techniques for Data Mining.

Satish Mandavalli is a Software Engineer at Microsoft with over twenty years of experience across finance, banking, healthcare, and enterprise IT systems. As a Chartered Accountant, he uniquely bridges finance and technology to design intelligent, data-driven solutions. His work focuses on applying AI and machine learning to optimize financial processes, improve risk management, and enable predictive analytics in real-world business environments. Satish is deeply invested in integrating smart, secure, and scalable AI-driven systems into financial operations. He is a strong advocate for innovation and continues to explore emerging technologies that enhance transparency, efficiency, and decision-making in financial and critical systems.

Shantanu Kumar is a Senior Software Engineer at Amazon, where he has been instrumental in developing and scaling key e-commerce initiatives since 2016. He played a central role in building and expanding Buy with Prime, enabling seamless integrations with Shopify, BigCommerce, Salesforce, and Meta platforms. His expertise spans scalable API design, secure checkout systems, and data-driven orchestration engines that have contributed significantly to Amazon’s global commerce ecosystem. He has also worked on machine learning–based recommendation systems for Prime Video and modernized largescale data ingestion pipelines. Shantanu is a recognized mentor and leader, ranked among the top 1% mentors on ADP List. He has conducted over 50 professional development sessions worldwide and has guided hundreds of professionals in career growth and technical leadership. A graduate of the National Institute of Technology (NIT) Kurukshetra, Shantanu has received multiple performance awards at Amazon and continues to drive innovation in scalable, intelligent systems.

Gautam Siddharth Kashyap is a Ph.D. researcher at Macquarie University, specializing in the alignment of Large Language Models (LLMs) via the HHH framework―Helpfulness, Harmlessness, and Honesty. His doctoral research focuses on developing principled methods to align LLMs with human values, leading to publications at EMNLP, EACL, etc. Beyond his doctoral research, Gautam also contributes to NLP for social good, focusing on the development of ethical, inclusive, and reliable AI systems.