Combining the unprecedented levels of data generated in the modern world, the use of high performance computing (e.g. distributed computing, GPUs/TPUs, etc.) and new innovations such as deep learning, AI has reached something of a golden age in recent years. However, for many organisations the challenge has increasingly become more how to practically implement these methods and to maximise the opportunities available. 

We seek papers that clearly address specific applications, develop innovative methods/methodologies and demonstrate real-world implementations. 

We welcome talks on many aspects of artificial intelligence, including:

•  AI methods and algorithms

•  AI case studies and applications

•  Data management for AI

•  Machine learning engineering and MLOps

•  Building an AI driven culture

•  Intersections of AI and OR

Modellers and researchers should find this stream particularly illuminating. The BOR stream welcomes contributions from both academics and practitioners.  

.

The stream aims to bring together researchers and practitioners to share their knowledge and experience in tackling combinatorial optimisation problems. It focusses on advances in algorithmic techniques for solving combinatorial optimisation problems of any kind, ranging from, e.g., cutting & packing, routing and scheduling problems to newer application areas such as computational biology and machine learning.

Contributions in this stream can range from the design of and analysis of algorithms to the derivation of structural properties of a problem, its models and its solutions. Computational work and novel applications to practical problems are especially welcome.From a methodological perspective, topics may include (but are not limited to):

• approximation algorithms and heuristic algorithms

• branch-and-bound, branch-and-cut and cutting-plane methods

• branch-and-price and column-generation methods

• decomposition methods, including Benders decomposition

• polyhedral combinatorics

• computational complexity

• on-line algorithms

Application areas include (but are not limited to):

• energy (production, storage, transmission)

• finance

• logistics

• manufacturing

• supply chain

• routing and transportation

• machine learning

. Building such models can be useful to identify and solve complex challenges within communities. They might also be effective to forecast and prepare communities for future trials and previously unforeseen events. Such models are usually effective in identifying and organising the internal and external resources accessible to communities, but their capability to develop new resources within such collectives is not always so evident. A relevant endeavour, therefore, is to explore and test ways to build models that deal with increasingly complex challenges by managing current variety levels and developing new abilities and resources within communities.

Traditional descriptions for such kind of complex challenges have been clearly bounded and exemplified (e.g., Arrow's impossibility theorem, Hardin's Commons problem, Axelrod's Prisoners' dilemma). To deal with these, different OR (soft) approaches have been proposed; some of them useful to increase individuals’ expertise and to overcome community barriers for development. However, supporting decision-making, problem-solving and improving operations in loosely-configured groups of people, who act and interact to achieve (sometimes conflicting) commitments, interests, and resources, is not a trivial task. The role that Community OR researchers may play is then relevant and the participation of community actors fundamental. Discovering how different players may support such communities is one of the general aims of Community OR. 

We would like to invite thoughts, contributions, and examples of any systematic ways to identify, organise and develop internal and external resources that can support communities facing increasingly complex challenges. Topics related to this include methods that support individual and collective resilience, sustainable collective development, etc. Contributions from academics and practitioners are equally welcomed.

We will be discussing the works around continuous optimisation and its interactions with other fields, such as machine learning and engineering, as well as relevant tools and methodological techniques, including first and second order methods, and derivative free approaches.

Both parametric and non-parametric papers will be considered. We especially welcome papers on the theory, methodology and application of Data Envelopment Analysis and econometric methods in performance management. Of particular interest are successful applications of performance and efficiency analysis in the real world, for example in banking, energy, environment,  healthcare, education, transportation, and so on.

Talks in this area may cover a broad range of topics, such as capacity planning, patient/appointment scheduling, waiting list management, care/treatment pathways, or decision problems related to the COVID-19 pandemic; but is clearly not limited to those. Applications of state-of-the-art or development of novel research methods using simulation paradigms, optimisation approaches, forecasting models, or artificial intelligence will showcase how best to address topical decision problems in the healthcare sector.

This stream will consist of talks from both theoretical and practical perspectives. Hence, attending talks and sessions will be valuable to academics, consultants, and practitioners alike.

This is a fast-growing research area with many communities actively involved, including but not limited to statistical ranking and selection, multi-armed bandit, optimal bidding, optimal stopping, kriging-based optimization, efficient global optimization, and learning in stochastic optimization. We encourage and invite papers on these broad areas. The stream welcomes presentations devoted to methodological, theoretical, or even empirical contributions.

It deals with the theory and application of stochastic programming and robust optimisation and includes methodologies that lie within their intersection, such as robust stochastic optimisation, distributionally robust optimisation and other data-driven or ambiguity-based approaches. 

The third sector includes a wide range of organisational forms, including charities, cooperatives, community groups and voluntary societies, each with their own governance arrangements and with a diverse range of skills, operating models, activities and beneficiaries.  As such decisions and performance need to be judged against a wide range of criteria, often with levels of complexity.  It therefore can present unique challenges and opportunities for analysts.

Analytical activity may include modelling for efficiency and effectiveness improvement; data analysis for increased insight, better forecasting or service design; benefits, outcomes and performance measurement; strategy development; and more.

The OR Society has a strategic objective to increase the number of organisations that benefit from OR.  Thus the OR Society's Third Sector Initiative, including the Third Sector Special Interest Group (SIG) and the OR Pro Bono Scheme, is aimed at improving OR visibility and impact within the sector.  The Pro Bono OR scheme connects volunteer analysts with good causes – thereby promoting awareness and understanding of the benefits of OR across the third sector (and to wider audiences) and giving OR analysts an opportunity to practise in a wider arena and develop their knowledge and skills.  These activities (of the OR Society’s Third Sector Initiative) bring OR support to organisations that have no analytical expertise.  They also provide a forum for those already working in OR in the third sector to:  firstly, share and develop good practice; and, secondly, establish relationships with kindred analysts working in the sector - whether data scientists, statisticians or strategists.  

The aim of this stream is to showcase any relevant analytical work, share experiences, and explore issues that may help improve the effectiveness of third sector OR and analysts.  Submissions may be case-study papers; discussions of work in progress or planned; papers discussing need or demand for analysis in relevant areas; and papers exploring the general issues associated with the practice of OR in the Third Sector.

Modes of delivery (e.g., hybrid, HyFlex, etc.); methods for evaluating teaching effectiveness, assessing authentically, and giving feedback effectively; challenge-based learning (e.g., hackathons, competitions,  industry challenges, role-play simulations, etc.); using data to personalise learning; enabling the student voice; techniques for engaging and enthusing students; running LMSs as fully interactive platforms; etc.

RL is a type of semi-supervised machine learning technique that enable an agent to learn in an interactive environment by trial and error using rewards to make sequential decisions under uncertainty. Applications of RL are widely used in the real world, including manufacturing, healthcare, trading, robotics and so on.This stream provides a fantastic opportunity to bring researchers from both OR and AI communities together to learn from each other and covers the topics of state of the art of RL theory, advanced RL algorithm, and applications.

We focus on all aspects of applied stochastic processes, risk, reliability and maintenance, which may be either applied or theoretical. Submissions on related topics including cases studies and new developments are also welcome.

Many companies are challenged by the desire to match the resources needed to satisfy the demand for their goods and services. Similarly using data to be able to configure the direction of a retail operation over the next 5 years is a hard problem - being, as it often is, a complex calculation around external economic and demographic trends.

The Retail Optimisation stream at OR64 welcomes and includes any presentation around use of optimisation and data science in a broad retail environment, examples include but are certainly not limited to:

* design and optimisation of an online fulfilment centre

* how to optimally spend marketing budget in a domain to maximise return

* optimal asset management around delivery and order fulfilment needed to service forecast demand

* optimal people scheduling to maximise happiness or flexibility of operation while servicing required orders

* having a framework for evaluating which retailers can optimally occupy a physical space

* deciding how to optimally price a scarce time-limited product on a website to maximise revenue

* physical store layout design to maximise sales

The topics can be theoretically focussed, computationally oriented or application based. The stream will provide a forum for both academics and practitioners in the fields of revenue management and pricing analytics. Talks on data-driven revenue management and pricing applications are strongly encouraged. 

. The stream will focus on methodological insights on the theory, methodology and practice of Soft OR and PSMs. Contributions may be to the development of a specific approach, technique or look across the field of PSMs to say something about them in general. Insights may be derived from case studies, action research programmes, experiments, secondary data or conceptual papers. For those seeking to publish work the stream organisers will seek to provide constructive feedback to help towards a successful review. 

This stream is interested both in modelling contributions (optimising end-to-end supply chain operations) and in empirical research (looking at the impact of the implementation of sustainable practices in supply chains, along with drivers and barriers).