EPSRC is the U.K.'s "leading funding agency for research and training in engineering and the physical sciences" (www.epsrc.ac.uk) and hence is Britain's broad equivalent of the National Science Foundation (NSF). Since 2003, EPSRC has, along with other U.K. funding agencies, been overseeing a program (under the title "Science and Innovation") of targeted investment in strategic areas of scientific and engineering research in order to "ensure the future international standing of the U.K. research base." O.R. was identified as such a strategic area for the 2007 Science and Innovation exercise. The document soliciting bids described the challenges facing O.R. in the U.K. in the following terms:

"The discipline of Operational Research (O.R.) uses advanced analytical methods quantitatively and qualitatively to improve the efficiency of operations or processes. These could be for example by achieving greater efficiency, better customer service, higher quality or lower cost. It provides generic methodologies that can be applied to a wide range of applications, many of which are strategically important to the U.K.

"The U.K. O.R. community is renowned internationally for the close linkage between academic research and applications. In order to maintain this reputation, it is vital that the core mathematical underpinning of the subject is nurtured to ensure that innovative methodologies continue to be developed. However, the review of the research status of O.R. in the U.K. highlighted concerns in the age distribution of academic O.R. researchers and the scarcity of students wishing to take Ph.D.s. There is a pressing need to bring in young researchers and to generate capacity in the fundamental aspects of O.R. The specific focus in this Science and Innovation topic is to grow capacity in O.R. that has a strong mathematical content with clear links to applications (their emphasis)."

The research council's analysis of the state of O.R. research in the U.K. had its roots in an international review of the subject that the council commissioned in 2004 and to which some eminent U.S.-based academics contributed. That review acknowledged the special role of the U.K. in the establishment and development of O.R. and its continuing maintenance of an application-oriented research tradition that is both highly distinctive and true to its roots. That said, the review also expressed to EPSRC the view that the U.K.'s leading position in applied O.R. could well be threatened by the dearth of U.K.-based researchers developing "new theory and methodology" in O.R. Furthermore, the age distribution suggested that, in the absence of appropriate action, things were not about to get better. The opinion of the international review team was that this situation constituted a real risk to the health of O.R. in the U.K. in the longer term. The $26 million initiative was awarded with the goal of addressing this risk.

The LANCS Initiative


When the successful proposals in EPSRC's 2007 Science and Innovation round were announced late last year, for the first time one of the grants had been awarded in O.R. The O.R. bid goes under the title "The LANCS Initiative in Foundational O.R.: Building Theory for Practice." This initiative will seek over the next five years to deploy the very considerable investment available to it to establish a major national hub of innovative research activity with the adventurous goal of closing the gap between applied O.R. and the theoretical issues that underpin it. The additional research capacity in theoretical O.R., which is the prime goal of the initiative, will be secured in the context of this inter-institutional research program that will embrace a multi-disciplinary agenda and actively seek to build upon an already extensive network of research and industrial collaboration in the U.K. and abroad.

Specific aims of the LANCS Initiative include:

• A significant enhancement of research leadership in O.R. within the U.K. through the recruitment of senior scientists of international standing. Additionally, the development of a strengthened pool of more junior academic staff. The academic recruitment program will aim to build U.K. capacity by attracting international candidates into the country to engage in this ambitious cross-institutional initiative.
  
  
• A building of U.K. O.R. capacity in a new and exciting way by establishing a joint research strategy across the four participating institutions.
  
  
• The development of a major program of international outreach to support collaborations with colleagues from around the world. The goal is to complement the initiative's national capacity-building strategy in order to make a major contribution toward setting the international agenda at the leading edge of work on the theoretical foundations of applied O.R.
  
  
• The development of a new generation of outstanding O.R. scientists in the U.K., in part by creating non-traditional routes into Ph.D. and postdoctoral positions by drawing upon inter-disciplinary synergies with other areas.
  
  
• Long-term sustainability of the additional research capacity achieved in the initiative.
  
  
• A closing of the gap between real-world inspired foundational O.R. and applications through a broad industrial outreach program that will engage closely with the U.K.'s Smith Institute in Industrial Mathematics and will draw upon a wide-ranging network of international industrial collaboration and affiliation.
  
  
• Prioritising knowledge transfer activities at the interface of O.R. theory and practice.

Green logistics. This topical and wide-ranging program of activity is concerned with the challenges posed by the production and distribution of goods in a sustainable way, taking account of environmental and social factors. Increasing interest in green issues from industry, government and the general public ensures that this will remain a key social concern for the foreseeable future. New challenges are posed by the adoption of new processes, such as developments in "reverse logistics," where recycled materials may be used at different stages of the production process. Waste management is another area where the adoption of recycling policies for household waste collection gives rise to new logistics management issues.

Green logistics is an application area for O.R. that is full of new, complex multi-objective problems. The research program will seek to develop innovative O.R. approaches that will enable a deeper understanding of environmental risk factors and the associated costs, benefits and trade-offs that will lead to more informed and more environmentally friendly decision-making.

Heuristic understanding. The last 40 years have seen a major international research effort in the design of heuristic methods to underpin decision-support system development. Thousands of research papers have been published over the years. However, our theoretical understanding of which methods work well in which situations is meager. The area has been characterised by the "trial-and-error" development of algorithms with little understanding of the corresponding search space landscape. The landscape is linked to an algorithm's ability to solve a problem, but current research in this area ignores this observation.

We aim to undertake a sustained effort in the mathematical analysis of fitness landscapes and algorithm performance with the goal of better informing heuristic algorithm design. Similar issues arise in the modeling, analysis and optimization of complex stochastic systems, which are central to a wide range of applications in computer and communication networks and in manufacturing and service systems. However, stochastic (dynamic) optimization in the context of such systems is a major technical challenge. A range of heuristic approaches have emerged (including heavy traffic, approximate dynamic programming, polyhedral methods and index policies) that have proved effective in specific contexts. However, there is as yet insufficient underlying theory offering general insights concerning the differential effectiveness of alternative approaches.

Systems to build systems. The current state of the art in the development of decision-support system methodologies is focused around the human design of bespoke systems, which are specifically tailored to the particular problem-solving environment in hand. This theme will draw upon an emerging area of research at the interface of O.R. and computer science that aims to underpin the development of automated systems to build and design search methodologies.

The goal is to (at least partially) remove the human from the process of designing computational search methodologies for real-world problems. This is a particularly adventurous research challenge that is being addressed in the (almost) complete absence of a mathematical and theoretical understanding of how to build intelligent systems that are capable of automatically building new systems. This initiative will establish a major cross-institutional effort to explore such issues. A deeper understanding of how and why heuristics work would feed into and inform the development of systems that can automatically and intelligently build and select search methodologies.

The interface between discrete and convex optimization. Discrete (or combinatorial) optimization is concerned with search over discrete structures, whereas convex optimization is concerned with continuous problems in which the objective and constraints are convex. Due to their many applications, both topics have received considerable attention from researchers, to the extent that they have each become fields in their own right. A practical outcome of this research effort has been the development of robust software packages that are capable of solving many large-scale problems of practical interest.

However, research in the two fields has taken place more or less independently. As a result, discrete convex optimization, a generalization of the two, is in its infancy. Yet, it has important applications, for example in facility location, resource allocation, machine scheduling, portfolio selection, statistical clustering and in finding "robust" solutions to certain discrete stochastic optimization problems. The planned research will be concerned with heuristics, bounding procedures and exact algorithms for this challenging class of problems.

Health care. O.R. has a major role to play in health care both in administrative logistics and in effectively and efficiently providing treatment. Part of the challenge in exploring the formidable mathematical issues that result is the need for solutions that are implementable in real-life scenarios. Many health care issues generate extremely complex O.R. problems or, more accurately, a complex interaction of several complex O.R. problems.

One of the key challenges we face is how to develop mathematical models and search methodologies that capture the complexity and uncertainty of modern health care environments. The area is characterized by scientifically demanding real-world problems that require a broad interdisciplinary approach and close collaboration with practitioners. Examples include controlling a potential avian flu pandemic, planning the future health care workforce in the context of an ageing population, scheduling patient pathways, controlling the escalating costs of long-term conditions such as obesity, diabetes and heart disease, managing global supply chain logistics for pharmaceutical and medical products, designing mass vaccination programs in the event of bioterrorism and optimizing the use of new non-invasive technology in cancer treatment. The list is endless.

Health care is arguably one of the areas where the most challenging and most difficult O.R. problems arise, yet compared with fields such as defense or manufacturing industry there has been relatively little take-up by practitioners over the years, despite a long tradition of academic research. The aim in this initiative is to reverse this trend by undertaking a far-reaching program of multi-disciplinary research at the foundations of health care O.R. that draws upon a wide network of collaboration with clinicians, managers and other health care professionals.

Transport. The efficient operation of transport systems involving air, train and bus travel has traditionally presented operational researchers with a variety of important and scientifically challenging problems. These include the design of routes, construction of timetables, assignment of duties to crew and drivers, and scheduling of airport runways and rail track. However, there is a need to build and analyze more complex models and solution methodologies as real-world transport scenarios evolve and adapt to societal changes.

An example of the type of challenging issue that will be addressed in this initiative is the dynamic routing of buses in response to customer demand. Another example is the design of robust routes and timetables that are less sensitive to disruptions that may be caused by breakdowns or bad weather, and algorithms that automatically reschedule when a disruption has occurred. One of the generic features of many of the problems we face is developing robust methodologies that can handle the conflicting objectives of satisfying customer demand in a timely way while minimizing the costs of the running the transportation service.

Opportunities for Engagement


We were keen to bring this major $26 million initiative to the attention of the INFORMS membership, not simply to share our understandable sense of excitement at the prospect of significant new research investment to take forward the subject we love in our own country, but also because there will be many opportunities for direct engagement with its program. For example:

• There will be academic positions to be filled at both senior and starting levels. For example, there will be full professorships, lecturers (equivalent to assistant professorships), post-doctoral research fellowships and Ph.D. studentships. Given that our primary goal is to grow national U.K. research capacity, we anticipate that much of our academic recruitment will be from overseas.
  
  
• There will be myriad research training opportunities at both doctoral and postdoctoral levels with ample opportunity for working with and learning from a broad community of academics, industrialists and practitioners.
  
  
• There will be a broad range of well-resourced opportunities for collaborative engagement within our extensive research program. Indeed, we have outlined an extensive program of international scientific and industrial outreach and collaboration.

The official start date for the initiative is Sept. 1, 2008, but a great deal of work will be in train well before then. Please keep an eye out for the first tranche of academic positions and research training opportunities to be advertised in the next few months. Indeed, you will see an advertisement for a range of positions elsewhere in this issue. Moreover, if you are not looking to move your career to the U.K. but are still interested in collaboration with the initiative, then a wide variety of opportunities will be available over the next few years.



* Contributors to this article include Kevin Glazebrook (director, LANCS initiative; k.glazebrook@lancaster.ac.uk), Richard Eglese, Adam Letchford (Lancaster University), Edmund Burke (University of Nottingham), Jeff Griffiths (Cardiff University) and Chris Potts (University of Southampton).

Table of Contents

OR/MS Today Home Page


OR/MS Today copyright © 2008 by the Institute for Operations Research and the Management Sciences. All rights reserved.


Lionheart Publishing, Inc.
506 Roswell Rd., Suite 220, Marietta, GA 30060 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com


Web Site © Copyright 2008 by Lionheart Publishing, Inc. All rights reserved.