https://onlinelibrary.wiley.com/page/journal/14753995/homepage/call_for_special_issue_papers.htm
https://onlinelibrary.wiley.com/pb-assets/assets/14753995/cfp/itor70148-integration-on-operations-cfp-1767626756223.pdf
Guest Editors:
The International Transactions in Operational Research (ITOR), the flagship journal published by the International Federation of Operational Research Societies, will publish a special issue dedicated to the Integration of Operations Research with Digital Twins and AI.
Digital twins (DT) and artificial intelligence (AI) are transforming operations research (OR). OR methods depend on the collection, processing, reasoning, and integration of data. While some of these tasks still require human expertise, many have become automated or can be managed by AI, such as data-driven and language-based model development. Historically, limitations in data availability and certainty led to the creation of OR methods based on probabilistic predictions. With DT and AI, these constraints are eased, prompting critical questions about the continued relevance and effectiveness of traditional optimization and simulation techniques for modern, AIdriven operations and supply chains. At the same time, the integration of OR with agentic AI ecosystems is opening new opportunities, which this special issue seeks to explore, together with potential challenges.
DT and AI fundamentally redefine what constitutes a “model.” Models can become dynamic, able to reconfigure their structure and functionality in response to data, AI-driven insights, and collaboration between humans and AI. This shift moves the focus from purely algorithmic engineering to creative engagement with AI assistants, evaluating agent interactions, validating optimization suggestions, and testing scenarios using real-time or frequently updated data under various uncertainties. Instead of optimizing isolated models, agentic AI and DT enable coordinated decision systems built on human-AI collaboration. Integrating with DT and AI is essential for OR to remain relevant in a data- and AI-centric landscape. Traditional optimization methods often overlook the needs of end users, who face complex, unstructured problems with multiple objectives and uncertainties. Users expect a flexible environment that enables real-time exploration of alternatives, scenarios, and uncertainties, and that allows them to adjust inputs and assumptions as needed. Stand-alone, offline models cannot deliver these real-time insights unless connected to external data sources and digital ecosystems.
OR models must use up-to-date, real-world data to remain practical. Creativity and adaptability are crucial as decision-support models evolve in fast-paced settings. Many OR models lack the capacity to learn and improve as data and environments change. Beyond adjusting parameters, structural changes—like adding constraints—may be needed to keep models relevant. Rapid detection of discrepancies, real-time updates, and model adaptation are critical for practical applications. Additionally, decision-makers often struggle to interpret OR model results, which can be complex and unintuitive. Unlike experts, managers typically use less formal, more verbal reasoning. Interpretability is a two-way street: humans must understand machine outputs, and machines must capture human intent. Recent literature has discussed interfaces of OR, DT, and AI in different forms, for example, supply chain DT, combining generative AI and optimization, integrating control theory and multiagent systems, and using AI and computer science methods for solving manufacturing and supply chain problems. That said, we are not aware of any special issue in a leading international journal addressing this novel and rapidly developing field.
In this special issue, we call for papers seeking to advance state-of-the-art and practical applications of OR methods in DTs and AI.
We aim at collating research combining OR methods with AI and DTs from different perspectives, including cognitive and intelligent DT, generative AI, AIbased analytics (e.g., machine learning, deep learning, and reinforcement learning), agentic AI, and multi-agent systems, for example:
With this respect, we would like to stress that ITOR is an OR journal, and we expect to receive submissions that contribute (also) to OR, not (only) to AI.
All papers submitted to this special issue will undergo the standard peer review procedures established by ITOR. Submitted papers must be original, unpublished, and not currently under consideration for publication elsewhere.
All submissions must fit within the journal’s domain statement and will be judged for their relevance to the special issue’s scope, innovativeness, and the extent of theoretical and practical research contribution. Contributions should be prepared according to the instructions to authors, available at the journal homepage on https://onlinelibrary.wiley.com/page/journal/14753995/homepage/forauthors.html.
Authors should submit and upload their contributions using the submission site https://wiley.atyponrex. com/journal/ITOR, indicating in their cover letter that the paper is intended for this special issue.
Other inquiries should be sent directly to the guest editors in charge of this issue:
Dmitry Ivanov (divanov@hwr-berlin.de);
Tsan-Ming (Jason) Choi (T.M.Choi@liverpool.ac.uk);
Seyyed Ehsan Hashemi-Petroodi (seyyed-ehsan.hashemi-petroodi@kedgebs.com); and
Eric K.H. Leung (K.H.Leung@liverpool.ac.uk).