Sridar will discuss two original approaches to solve nonlinear integer optimisation problems that arise in applications in finance, cancer genomics and supply chain optimisation. The Graver Augmented Multiseed Algorithm (GAMA) utilises augmentation along Graver basis elements (the improvement direction is obtained by comparing objective function values) from these multiple initial feasible solutions.
- A hybrid quantum classical approach (GAMMA-Q) that have the potential to solve a variety of hard linear and nonlinear integer programmes, as the form a test set (optimality cerficate). We test two hybrid quantum classical algorithms (on D-Wave) one for computing Graver basis and a second for optimising nonlinear integer programs that utilise Graver bases to understand the strengths and limitations of the practical quantum annealers available today. Our experiments suggest that with a modest increase in coupler precision along with near term improvements in the number of qubits and connectivity that are expected the ability to outperform classical best in class algorithms is within reach.
- A (fully classical) approach (GAMA-C) to solving certain non-convex integer programmes. This method is well suited for Cardinality Boolean Quadratic Problems (CBQP), Quadratic Semi Assignment Problems (QSAP) and Quadratic Assignment Problems (QAP). Sensitivity analysis indicates that the rate at which GAMA slows down as the problem size increases is much lower than that of Gurobi. We find that for several instances of practical relevance, GAMA vastly outperforms in terms of time to find the optimal solution (by two or three orders of magnitude).
- Results of applying GAMA on data from The Cancer Genome Project (TCGA) to find mutated driver pathways are encouraging. I will discuss some results on Acute Myleoid Luekemia (AML) and Glioblastoma Multiforme (GBM).
Sridhar Tayur is the Ford Distinguished Research Chair and University Professor of Operations Management at Carnegie Mellon University’s Tepper School of Business. He received his PhD in Operations Research and Industrial Engineering from Cornell University and his undergraduate degree in Mechanical Engineering from the Indian Institute of Technology (IIT) at Madras (where he is a Distinguished Alumnus Award winner). He is an INFORMS Fellow, a Distinguished Fellow of MSOM Society and has been elected to the National Academy of Engineering (NAE).
He has been a visiting professor at Cornell, MIT and Stanford. He has published in Operations Research, Management Science, Mathematics of Operations Research, Mathematical Programming, Stochastic Models, Queuing Systems, Transportation Science, POMS, IIE Transactions, NRLQ, Journal of Algorithms and MSOM Journal. He has served on the editorial boards of Operations Research, MSOM Journal, Management Science, IIE Transactions and POMS. He served as President of MSOM Society.
He has co-edited “Quantitative models for supply chain management” (1998) and “Handbook of healthcare analytics” (2018).
He has been a finalist for the Lanchester Prize and is an Edelman Laureate. He has won the Healthcare Best paper Award by POMS and the INFORMS Pierskalla Award for best paper in Healthcare. He has won the Gerald L Thompson Teaching Award in the BS Business Administration Program, the George Leland Bach Excellence in Teaching Award given by MBA students, the INFORMS Teaching Case award, and has been named as a “Top Professor” by Business Week.
Previous research has shown that placing patients in non-dedicated medical units increases patient length of stay (LOS) and mortality. However, in many cases, the “optimal” LOS and the best location for treatment of the patient is not obvious. A case in point are haematology patients, for whom this is a critical decision. These patients are hospitalised on a regular basis for chemotherapy treatments and it is debated whether these treatments should be offered during hospitalisation followed by an observation period, or whether these patients are better treated in outpatient clinics and should stay at home unless infection is suspected.
Patients with haematological malignancies are susceptible to life-threatening infections after chemotherapy. Hence, LOS optimisation of haematology patients must balance the risk of patient infection with mortality. The former is minimised by minimising hospital stay, while the latter is reduced through hospital care, where infections are identified and treated earlier.
We analyse longitudinal patient data of a large tertiary hospital and evaluate whether management of such patients in dedicated inpatient and emergency wards could provide superior infection prevention and better outcomes. Electronic medical records were analysed to retrieve infection related information and on this basis, patient-tailored risk prediction models were developed. These models help us to empirically compare different patient management policies.
We then developed a Markov Decision Process formulation to explore the connection between these non-monotonic risk functions and the optimal LOS from a single patient perspective. We further consider the social problem in which capacity constraints limit the ability of hospitals to keep patients for their optimal LOS. We find that the optimal policy under this constraint takes the form of a two threshold policy. This policy either blocks some of the patients and immediately routes them to home care observation (under the outpatient clinic guidance), or speeds up some of the patients and routes them to home care after an observation period in the hospital.
Our optimisation model assumes general hazard rate functions, and therefore, can capture the changing dynamics inpatient health through time. It can help guide physician decisions in real-time, as patient state evolves.
Galit Yom-Tov is an Assistant Professor at IE&M faculty of the Technion. Her research focuses on operations of service systems, in particular, healthcare and contact centres. Dr Yom-Tov is the Co-Director of the Service Engineering Enterprise Lab (SEE-Lab) a worldwide hub for research and teaching in Service Engineering. Her research aims to build models for understanding the impact of customer and agent behaviour on service systems and to incorporate these behaviours into operational models of such systems. Her multidisciplinary research approach applies a combination of data science and stochastic modelling to archives of digital traces from service systems. Her recent work used such data to study the dynamics of customer emotions in contact centres, the reaction of customers to waiting announcements in emergency departments, as well as other aspects of service delivery. Dr Yom-Tov has published her work in leading operations research journals, including Management Science, Manufacturing & Service Operations Management, Operations Research, and Stochastic Systems.
Social media has emerged as an important channel to disseminate quality information to consumers in a variety of service settings, ranging from hotels to restaurants to home repairs. Its presence has recently spread to healthcare services, where government report cards have long been established to disclose quality information to the public. Although the effects of social media and government report cards on consumer choice have been previously studied, they are investigated in separate settings, and little work exists to study how they interact with each other in influencing consumer choice. We seek to address this question in the context of the US nursing home industry by quantifying the joint effect of online consumer ratings from Yelp and government ratings from the Centers for Medicare and Medicaid Services. Both quality rating systems adopt a five star rating scale, making them comparable to each other from consumers’ perspective. We apply the difference in differences method with instrumental variables to estimate the effects of Yelp and government ratings on nursing homes’ resident admissions. We find that Yelp ratings may contradict government ratings. Nevertheless, higher Yelp ratings led to higher resident admissions on average. This effect was not uniformly distributed across consumers with different payment sources: Admissions of lucrative consumers, for example, Medicare covered residents, increased, whereas admissions of economically disadvantaged consumers, such as, Medicaid covered residents, decreased. Compared to government ratings, Yelp ratings exerted a stronger effect on consumer choice. Specifically, an additional star in the Yelp rating would on average increase a nursing home’s Medicare admissions by two percent more than an additional star in the government rating. To explain this result, we propose two mechanisms and provide supporting empirical evidence. First, Yelp’s brand recognition in other service settings helps to increase consumer awareness of its nursing home ratings, which we call the brand spillover effect. Second, the narrative presentation of consumer reviews on Yelp’s platform enables consumers to better process quality information, which we call the cognitive empathy effect. We also find that an additional star in the Yelp rating would on average increase a nursing home’s net income by $60,951 and total margin by 0.005. Although it appears to be financially beneficial, we find little evidence that nursing homes attempted to improve their Yelp ratings.
Dr Susan F. Lu is the Gerald Lyles Rising Star Associate Professor of Management at the Krannert School of Management, Purdue University. She received her PhD from the Kellogg School of Management at Northwestern University. She is also an affiliated faculty of healthcare engineering at the Regenstrief Center for Healthcare Engineering. Her research centers on healthcare operations and analytics, with an emphasis on nursing home operations and cardiac care delivery. Applying both empirical and machine learning methodologies, she investigates the operational drivers of healthcare delivery performance to understand the impact of public policies and technological innovations on the management of healthcare operations.
Dr Lu has published 21 research papers, many of which appeared in leading management and economics journals such as Science, Nature Scientific Reports, Management Science, POMS, Review of Economics and Statistics and Journal of Health Economics. She received the best paper award from the ASHE in 2008, which is one of the most prestigious awards in the community of health economists and is given to a single-authored paper every other year. In 2014, she received the Early Career Investigator from the National Institute of Health (NIH-HMORN) conference. In 2015, her paper about health IT received the WHITE best paper award, the most prestigious award in the field of health IT. In 2016, her paper about mandatory overtime laws was a finalist for the Pierskalla award by INFORMS Health Applications Society. In 2018, her work on treatments of heart attack was selected as a Best Abstract of the CRT 2018 Cardiovascular Research Technology Conference, which is one of the world’s leading interventional cardiology conferences and is attended by more than 3,000 doctors. In 2019, her paper on telemedicine won the best paper award in the Hawaii international conference on system Science.
Dr Lu’s work has gained considerable visibility in the fields of management and economics. One of her work was selected by Nature News for annual important discoveries in 2013. One work collaborating with a group of interdisciplinary researchers has been endorsed by the world known blog Freakonomics. One work combining operations management into policy analysis is recommended by a healthcare media Healthcare Value Hub. Another work on solving shortage of donated blood is recognised by the Nobel Prize Laureate Al Roth’s blog – Market Designer. Her paper which applies machine learning based techniques into empirical research appeared in the NBER Digest and Vox. Moreover, her work on the online physician rating platform has received attention from Castlight Health, a listing company.
We study supply chains where multiple suppliers sell to multiple retailers through a wholesale market. In practice, we often observe that both suppliers and retailers tend to influence the wholesale market price retailers pay to suppliers. However, existing models of supply chain competition do not capture retailers’ influence on the wholesale price (for example, buyer power), and show that the wholesale price and the order quantity per retailer do not change with the number of retailers. To overcome this limitation, we develop a competition model based on the market game mechanism in which the wholesale price is determined based on both suppliers’ and retailers’ decisions. When taking into account retailers’ buyer power, we obtain the result that is consistent with the observed practice: as the number of retailers increases, each retailer’s buyer power decreases, and each retailer is willing to pay more for their order, so the wholesale price increases. In this case, supply chain expansion to include more retailers (or suppliers) turns out to be more beneficial in terms of supply chain efficiency than what the prior literature shows without considering buyer power. Finally, we analyse the integration of two local supply chains and show that, although the profit of the integrated supply chain is greater than the sum of total profits of local supply chains, integration may reduce the total profit of firms in a retailer oriented supply chain that has more retailers than suppliers.
Dr Gizem Korpeoglu is an assistant professor of Industrial Engineering at Bilkent University. Before joining Bilkent, she was a postdoctoral fellow of Operations and Technology at University College London, School of Management. She received her PhD in Economics from Tepper School of Business at Carnegie Mellon University, and also holds an MS degree from there. She received her BS degree from Middle East Technical University.
Dr Korpeoglu studies operational problems in traditional and innovative marketplaces using game theoretical models. During her PhD studies, she has worked on competition models and mechanism design. After her PhD, she has applied her methodological knowledge on markets and competition models to supply chains, crowdsourcing platforms, and other online marketplaces. Her work has appeared at leading operations journals such as Management Science and Manufacturing and Service Operations Management, and economics journals such as Journal of Mathematical Economics and Economic Theory. She is a member of the Institute for Operations Research and Management Science (INFORMS) and Production and Operations Management Society (POMS), and the recipient of the second prize at INFORMS Technology Innovation Management and Entrepreneurship Section (TIMES) best working paper competition in 2018.
In this seminar, Dr Ren will present two recent empirical papers on healthcare that are co-authored with his colleagues. One paper uses “small data”, for example, data collected within a hospital, to uncover the relationship between hospital internal operations and its quality outcome. The other paper uses “big data”, for example, a nationwide healthcare insurance claim data set, to study the cost implications of high deductible health plans.
Details on the two published papers are as follows: Zheng, S., Tucker A. L., Ren, Z.J., Heineke, J., McLaughlin, A., Podell, A. L. (2018) “The impact of internal service quality on preventable adverse events in hospitals.” Production and Operations Management 27(12):2201-2212
Zheng S., Ren, Z. J., Heineke, J., and Geissler, K. (2016) “Reductions in diagnostic imaging with high deductible health plans”. Medical Care 54(2):110-7 (Lead Article)
Z. Justin Ren is an Associate Professor of Business Administration at Boston University Questrom School of Business, and a faculty researcher at Boston University Institute of Sustainable Energy (ISE). He was also a Research Affiliate at Massachusetts Institute of Technology (MIT) Sloan School of Management (2009-2014).
At Boston University, Dr Ren teaches Core operations management courses and a data analytics course that helps managers gain market intelligence and make strategic decisions. He also teaches in executive education in financial risk management. He is a certified teacher by the Harvard Business School Case Method Discussion Leadership Program.
Dr Ren’s research focuses on supply chain coordination, healthcare service quality, and clean energy transition. His research has appeared in publications such as Management Science, Operations Research, Production and Operations Management, and Medical Care. He has received several recognitions, including the INFORMS George B. Dantzig Dissertation Award, INFORMS Junior Faculty Paper Competition Award, and the Production and Operations Management Society (POMS) Wickham Skinner Early-Career Research Accomplishments Award. His consulting clients include INTEL, Staples, BestBuy, Payless Rental Car, PWC, among others.
Professor Ren received his MA degree from the University of Wisconsin-Madison, his MS and PhD in Operations and Information Management from The Wharton School at the University of Pennsylvania.
Few issues in the healthcare ecosystem are more salient than the utilisation of medical tests. By some estimates, up to 30 per cent of medical testing decisions are deemed inappropriate, which may entail either over or under testing. All too frequently, the public attention has centred on over-testing. By comparison, under testing has received little media coverage, but is prevalent in the medical literature. In addition, contrary to popular belief, the US trails most OECD countries in terms of the utilisation of medical tests.
In this talk, Tinglong will discuss a series of modelling efforts aimed at understanding diagnostic experts’ decision making processes. These efforts, motivated by the interventional cardiology setting, seek to provide a theory of under testing by accounting for both reputational and revenue inducement incentives. Tinglong will also highlight implications for policymakers and healthcare executives with regard to incentive design for improving diagnostic accuracy.
Tinglong Dai is an Associate Professor of Operations Management and Business Analytics at Johns Hopkins University, Carey Business School. His research, recognised by many awards such as Johns Hopkins Discovery Award, INFORMS Public Sector Operations Research Best Paper, and POMS Best Healthcare Paper Award, spans across healthcare, marketing/operations interfaces, and AI-enabled business.
Tinglong’s research has been accepted for publication in leading journals such as Management Science, M&SOM, Marketing Science, Operations Research, and INFORMS Journal on Computing. He is an Associate Editor of Naval Research Logistics and is on the Editorial Review Board of Production and Operations Management. He co-chairs the Johns Hopkins Symposium on Healthcare Operations and co-edits the Handbook of Healthcare Analytics: Theoretical Minimum for Conducting 21st Century Research on Healthcare Operations, published by John Wiley & Sons in 2018.
Tinglong received his PhD (2013) and MS (2009) in Operations Management/Robotics from Tepper School of Business, Carnegie Mellon University, in addition to an MPhil (2006) in Industrial Engineering from the Hong Kong University of Science and Technology.
The Hospital Readmissions Reduction Program (HRRP) reduces Medicare payments to hospitals with higher than expected readmission rates where the expected readmission rate for each hospital is determined based on national average readmission levels.
Although similar relative performance based schemes are shown to lead to socially optimal outcomes in other settings, HRRP differs from these schemes in three respects:
- deviation from the targets are adjusted using a multiplier
- the total financial penalty for a hospital with higher-than-expected readmission rate is capped
- hospitals with lower-than-expected readmission rates do not receive bonus payments.
We study three regulatory schemes derived from HRRP to determine the impact of each feature, and use a principle-agent model to show that:
- HRRP over-penalises hospitals with excess readmissions because of the multiplier and its effect can be substantial
- having a penalty cap can curtail the effect of financial incentives and result in a no-equilibrium outcome when the cap is too low
- not allowing bonus payments leads to many alternative symmetric equilibria, including one where hospitals exert no effort to reduce readmissions.
These results show that HRRP does not provide the right incentives for hospitals to reduce readmissions.
Next we show that a bundled payment type reimbursement method, which reimburses hospitals once for each episode of care (including readmissions), leads to socially optimal cost and readmissions reduction efforts.
Finally we show that, when delays to accessing care are inevitable, the reimbursement schemes need to provide additional incentives for hospitals to invest sufficiently in capaci
Tolga Tezcan is a Professor of Management Science and Operations at London Business School (LBS). He teaches courses in data mining and business analytics. Prior to joining LBS, he was a faculty member at Simon School of Business in University of Rochester between 2010 and 2015, where he was placed in the teaching honour roll in 2014, and at University of Illinois at Urbana-Champaign between 2006 and 2010.
Tolga holds a PhD in industrial and systems engineering and a MS in mathematics from Georgia Tech, a MS in industrial and systems engineering from Colorado State-Pueblo, and a BS in industrial engineering from Bilkent University, Turkey. Tolga’s research focuses on the robust management of service systems, such as customer service centres and healthcare systems, under uncertainty.
His research has appeared in leading journals such as Management Science, Operations Research, M&SOM, and Annals of Applied Probability.
He has received the Career Award from National Science Foundation (NSF) of USA in 2010.
Most data-driven tools do not ask people for their input. We study whether and how to incorporate the discretion/expertise/intuition of physicians into data-driven decision support tools for improved operational decision-making in hospitals, in an empirical setting of predicting the surgery duration.
We consider three families of models for predicting the surgery duration: models with physician input, statistical models, and models that combine the two. Using the operating room scheduling and usage data from an academic hospital collected over three years, we empirically evaluate and compare the performances of the different models. We find that physician input offers predictive power beyond that of statistical models in our empirical setting. The best performing model is the combined model which includes the physician input as a feature in a statistical model with other predictors. The corrected physician input model, which applies a simple correction to the physician input, performs comparably well (the mean squared error increases by five per cent), when corrections are applied at the surgeon-procedure level. Without the physician input, the mean squared error of the best performing model increases by 17 per cent. We also compare the performances of the operating-room schedules resulting from our models and show that the results carry over.
Our findings suggest that hospital managers should consider eliciting physician input and incorporating it into data-driven decision support tools. We also show how physician input can be best leveraged in the context of predicting the surgery duration. Understanding the benefits/costs of allowing expert input in decision-making has been attracting the interest of the OM community in various application areas. We contribute to this line of research by studying the effect of incorporating physician input on operating room use. For hospital managers, we show the potential value of incorporating physician input in data-driven decision support tools.
This is joint work with Rouba Ibrahim (University College London).
Song-Hee Kim is an Assistant Professor of Operations Management in the Marshall School of Business at the University of Southern California. Her research interests are on using empirical and statistical analysis to build models for decision-making. She has a particular interest in healthcare applications, and has been actively involved in collaborative research projects with several hospitals including the Keck Hospital of USC, Yale New Haven Hospital, Kaiser Permanente in Northern California and Samsung Medical Centre in South Korea. Her research has been published in Management Science, Operations Research, and Manufacturing & Service Operations Management. She has received several academic awards including the Best OM Paper in Management Science Award (winner), MSOM Best Paper Award (finalist), INFORMS Pierskalla Award (finalist), MSOM Student Paper Award (first place), and INFORMS Health Applications Society Student Paper Award (finalist). Song-Hee completed her PhD in Operations Research from Columbia University and earned her BS in Operations Research and Industrial Engineering from Cornell University. Prior to joining the Marshall School, she was a postdoctoral associate at the Yale School of Management.
An ever-expanding array of disasters and emergent public health threats has raised operationally-relevant research questions about the willingness of public health and other healthcare workers to fulfil organisational expectations in disaster response and recovery. Dr Daniel Barnett from Johns Hopkins Bloomberg School of Public Health and colleagues have examined perceptual and attitudinal barriers and facilitators toward such willingness among these cohorts, including a focus on health organisational leadership perspectives in such contexts. Their findings have yielded novel, behavioural model-based curricular interventions to address these willingness gaps in health organisations. This presentation will accordingly focus on findings from mixed-methods research, including regarding these workers’ and their leaders’ commitment to, and sense of efficacy in the context of, their organisations in the face of disasters and emergent threats to public health and safety. The presentation will include currently unpublished data on health departments workers’ perceptions toward disaster recovery in a variety of Hurricane Sandy-impacted US jurisdictions from the states of Maryland and New Jersey. The presentation will also explicitly address broader organisational behaviour-relevant implications of these and related findings for current and future disaster preparedness, response and recovery efforts, including in international contexts.
Daniel Barnett, MD, MPH is an Associate Professor in the Department of Environmental Health & Engineering at the Johns Hopkins Bloomberg School of Public Health (JHSPH), where he has a joint appointment in the Department of Health Policy and Management and is on the Core Faculty of the Office of Public Health Practice & Training. His research interests include evidence-based approaches to organisational enhancement of public health emergency preparedness. Dr Barnett previously worked at Baltimore City Health Department’s Office of Public Health Preparedness and Response, where he conducted disaster preparedness training activities for the department’s workers. He received his MD degree from The Ohio State University College of Medicine and Public Health; his MPH degree was earned at The Johns Hopkins Bloomberg School of Public Health and he is a graduate of the Johns Hopkins General Preventive Medicine Residency Program.
Michaelmas Term 2018
In this presentation, Özalp will discuss when, how, and why the behavioural motives of trust and trustworthiness arise to support cooperation within and across businesses.
The session will identify four building blocks of trust and trustworthiness: personal values and norms, market environment, business infrastructure, and business process design. Özalp elaborate on these building blocks and offer tangible insights about how to establish trusting and cooperative business relationships. To do so, the session will provide a high level summary of some research results and case studies from across industries.
As part of the upcoming seminar, the group will play an interactive game. Please bring a laptop and also read the game instructions before the seminar (provided) – it takes only five minutes to do so. Faculty and students (participants of the seminar) will be playing with each other and the winner will get a prize. Participants will be encouraged to think a little about strategy (but don’t discuss it with others)… It will be a fun game!
Özalp Özer is Ashbel Smith Professor of Management Science at the University of Texas at Dallas, Jindal School of Management. He spent his 2013/14 sabbatical as a Visiting Professor at MIT Sloan School of Management. Previously he was a faculty member at Columbia University and Stanford University. His areas of specialty include end-to-end management and coordination of global value chains, strategic investment decisions, capacity and inventory planning, market timing, distribution channel management, procurement contract design, and retail and pricing management. Besides scuba diving, he is passionate about working with researchers and practitioners on the next new ‘think’ that calls for the exciting opportunity to explore, learn, and contribute. Özalp is also a recipient of the Wickham Skinner Early-Career Research Accomplishment Award from POM Society, the Hellman faculty fellowship, the Terman faculty fellowship, and the Eugene Grant Teaching Award at Stanford by vote of students in 2003 and 2004 and teaching awards at Columbia in 2009 and at MIT in 2014. PoetsandQuants.com announced him to be named as a Favorite Professor by Top Executive MBA programme students. Özalp is also an editor of The Oxford Handbook of Pricing Management published by Oxford Univerity Press in 2012. His articles have appeared in journals such as Management Science, and Operations Research. He is currently serving as an associate editor for Management Science, M&SOM, Operations Research, and Production and Operations Management. He is an active consultant to industry and has consulted companies such as General Motors, Hewlett Packard, Hitachi GST, IBM and Neiman Marcus. He received his PhD and MS degrees from Columbia University.
Easter Term 2018
The Patient Protection and Affordable Care Act (ACA) of 2010 represents one of the most significant regulatory overhauls of the United State healthcare system since the initial establishment of Medicare and Medicaid. While attention has largely been directed towards provisions that expanded medical coverage, the ACA also established several incentive schemes aimed at reforming the care delivery process by holding hospitals accountable for their performance in order to rein in costs and improve quality. One of the first such schemes introduced was the Hospital Readmissions Reduction Program (HRRP), which requires the Centres of Medicare and Medicaid Services (CMS) to reduce payments to hospitals that exhibit higher than average 30-day risk-adjusted readmission rates.
This paper examines the impact of the Hospital Readmissions Reduction Program (HRRP) on hospitals’ admission behaviour. We exploit variation in hospitals’ financial exposure to HRRP penalties due to i) readmission performance, and ii) financial constraints, to show that hospitals reduced readmissions post HRRP at least to some extent by increasing the number of patients that were classified as admitted for “observation.” Under this classification patients do not count as admissions for HRRP purposes. This increase is estimated to be 12.7 per cent more for hospitals that were exposed to HRRP penalties compared to non-penalised hospitals, and as much as 31.1 per cent if the hospital was also financially constrained.
The magnitude of this effect is operationally significant; rough calculations suggest that over 40 per cent of the readmission reduction that followed HRRP can be attributed to the change in observation bed usage. Our results also have implications for the implementation of HRRP which is based on average-performance benchmarks. When hospitals use observations beds to reduce their readmission figures, they also lower the average readmission rate against which other hospitals are penalised, hurting high-performing hospitals not managing readmissions though observation beds.
Research co-authored by Chris Chen.
Nicos Savva is an Associate Professor of Management Science and Operations at London Business School, where he teaches courses on data analytics, modelling, and healthcare management. Nicos’s research examines healthcare operations and innovation and has appeared in leading journals such as Management Science, Manufacturing & Service Operations Management, and Production and Operations Management. Nicos holds editorial positions in Management Science, Manufacturing & Service Operations Management, and Production and Operations Management. He holds a PhD in Management Science, an MPhil in Finance, and an MA in Natural Science (Physics), all from the University of Cambridge.
In many contexts such as product design and advertising, clients seek the expertise of external providers to generate innovative solutions for their business problems. In such delegated engagements, providers can improve the quality of solutions through the intensity of their efforts, and clients can evaluate solutions and decide when to stop the project. In this paper, we explore how the client’s flexibility in stopping the project influences the progress and efficiency of the delegated innovation. In particular, we compare two structures: “committed”, where the client stops the project immediately if the provider delivers an acceptable solution, and “open-ended”, where the client retains the flexibility to continue the project even after receiving an acceptable solution. We show that, when innovation is delegated, the client’s flexibility can lead to lower early efforts by the provider and thus may not always benefit the client. We generate insights regarding the appropriateness of the two structures with respect to the problem difficulty and provider’s capability. In addition, we extend our model and analysis in several directions by capturing the effects of client’s transparency, optimal payments, project timeline, and provider’s capability improvement.
Morvarid Rahmani is an Assistant Professor of Operations Management, at the Scheller College of Business, at Georgia Tech. She received her PhD from the UCLA Anderson School of Management. She also received three masters degrees, in Industrial Engineering, Electrical Engineering, and Economics.
Dr Rahmani’s research brings together the operational perspective of process improvement and the economic perspective of innovation and collaboration. Her research focuses on the study of the dynamics of collaboration in knowledge-based work processes such as new product or service development, management and IT consulting, technical projects, and education. Her research generates insights for advancing strategic decision-making, both across organisations and within them. She has published her research in Management Science, and Production and Operations Management journals. Her dissertation research paper on Collaborative Work Dynamics was a finalist in the Manufacturing & Service Operations Management Best Student Paper Competition.
Dr Rahmani has taught Core Operations Management in full-time and evening MBA programmes, and a seminar course on Managing Innovation and Product Development in the PhD programme at the Scheller College of Business. She has received the Brady Family Award for Faculty Teaching Excellence at the Scheller College of Business.
About one out of six inmates in the United States (US) is infected with hepatitis C virus (HCV). HCV prevalence in prison systems is 10 times higher than the general population, and hence prison systems offer a unique opportunity to control the HCV epidemic. New HCV treatment drugs are very effective, but providing treatment to all inmates is prohibitively expensive, which precludes universal HCV treatment in prison systems. As such, current practice recommends prioritising treatment based on clinical and incarceration-related factors, including disease staging, remaining sentence length, and injection drug use (IDU) status. However, there is controversy about how these factors should be incorporated because of the complicated tradeoffs.
In this study, we propose a restless bandit modelling framework to support hepatitis C treatment prioritisation decisions in US prisons. We first prove indexability for our problem and derive several structural properties of the well-known Whittle’s index, based on which, we derive a closed-form expression of the Whittle’s index for patients with advanced liver disease. From the interpretation of this closed-form expression, we anticipate that the performance of the Whittle’s index would degrade as the treatment capacity increases; and to address this limitation, we propose a capacity-adjusted closed-form index policy. We parameterise and validate our model using real-world data from Georgia state prison system and published studies. We test the performance of our proposed policy using a detailed, clinically-realistic simulation model and show that our proposed policy can significantly improve the overall effectiveness of the hepatitis C treatment programmes in prisons compared with the current practice and other benchmark policies, including the commonly used Whittle’s index policy.
Our results also shed light on several controversial health policy issues in hepatitis C treatment prioritisation in the prison setting and have important policy implications including: 1) prioritisation based on only liver health status, a commonly practiced policy, is suboptimal compared with many other policies we consider. Further, considering remaining sentence length of inmates and IDU status in addition to liver health status in prioritisation decisions can lead to a significant performance improvement; 2) the decision of whether to prioritise patients with shorter or longer remaining sentence lengths depends on the treatment capacities inside and outside the prison system, and prioritising patients with shorter remaining sentence lengths may be preferable in some cases, especially if the treatment capacity inside the prison system is not very tight and linkage-to-care level outside prison system is low; and 3) among patients with advanced liver disease, IDUs should not be prioritised unless their reinfection is very-well controlled. Lastly, we introduce and discuss a decision support tool we have developed for practical use.
Turgay Ayer is the George Family Foundation Early Career professor and an associate professor at Industrial and Systems Engineering, and is the research director for medical decision-making in the Center for Health & Humanitarian Systems at Georgia Tech. In addition, Dr Ayer has a courtesy appointment at Emory Medical School.
His research focuses on healthcare analytics, with applications in predictive health, medical decision making, healthcare operations, and health policy. His research papers have been published in top tier engineering, management, and medical journals, and covered by popular media outlets, including the Wall Street Journal, Washington Post, US News, and NPR.
Dr Ayer has received several awards for his work, including an NSF CAREER Award (2015), Society for Medical Decision Making (SMDM) Lee Lusted Award (2009), first place in the MSOM Best Practice-Based Research Competition (2017), and a finalist in the 2017 INFORMS Franz Edelman Competition (2017).
Ayer received a BS in industrial engineering from Sabanci University in Istanbul, Turkey, and his MS and PhD degrees in industrial and Systems Engineering from the University of Wisconsin – Madison.
Ayer is a member of INFORMS and Society for Medical Decision Making, an associate editor for Operations Research, and is a past president of the INFORMS Health Application Society.
Lent Term 2018
This study focuses on contracting for a three-tier supply chain consisting of a buyer, tier one supplier, and tier two sub-supplier where disruptions of random length occur at tier two. As is common in many supply chains, the buyer has a direct relationship with the tier one supplier but not the tier two sub-supplier; that is, the buyer has limited supply chain visibility. Both the supplier and sub-supplier can reserve emergency capacity prior to observing the disruption to protect the supply chain from the disruption. The study looks at how the buyer and the supplier can guarantee that the correct level of emergency capacity is built prior to the disruption. Due to two types of inefficiencies – a special form of double-marginalisation and the substitution effect – the supply chain is misaligned in its decentralised form. Although the lack of visibility prevents the buyer from directly contracting with the sub-supplier to eliminate these inefficiencies, they can still coordinate the supply chain through cascading, for example, contracting with the supplier, who in turn contracts with the sub-supplier. Despite supply chain coordination, the supplier benefits from the buyer’s limited supply chain visibility.
Collaborative work with Dr Georg Schorpp and Professor Hau Lee.
Dr Feryal Erhun is Reader in Operations Management Cambridge Judge Business School, University of Cambridge. Feryal’s research interests are in the topic of supply chain management, including risk management in supply chains, new product transitions, and supply contracts. Her current research also includes studies healthcare operations.
Dr Erhun is a strong proponent of practice-based research. In collaboration with Intel Corporation, her research group has designed a decision-support system for optimising capital investment decisions for firms in capital-intensive industries. This work has been selected as one of the finalists in the 2012 Edelman competition and Feryal has been inducted as an Edelman Laureate. She is also a recipient of 2006 NSF CAREER Award.
Feryal serves on the editorial boards of Manufacturing and Service Operations Management and Production and Operations Management. She received her PhD in Business Administration, with a concentration in Production and Operations Management from the Graduate School of Industrial Administration, Carnegie Mellon University in 2002. She holds a BS and a MS in Industrial Engineering from Bilkent University.
Replicating portfolios have emerged as an important tool in the life insurance industry, used for the valuation of companies’ liabilities. This paper describes the replicating portfolio (RP) model for approximating life insurance liabilities in place in a large global insurance company. We describe the challenges presented by the latest solvency regimes in Europe and how the RP model enables this company to comply with the Swiss Solvency Test. The model minimises the L1 error between the discounted life insurance liability cash flows and the discounted RP cash flows over a multi-period time horizon for a broad range of different future economic scenarios. A numerical application of the RP model to empirical data sets demonstrates that the model delivers RPs that match the liabilities and perform well for economic capital calculations.
Karl Schmedders has been a Professor of Quantitative Business Administration in the Faculty of Business, Economics, and Informatics at the University in Zurich since 2008. In addition, he is a Visiting Professor of Executive Education at the Kellogg School of Management, Northwestern University in Evanston, USA.
Karl received a PhD in operations research from Stanford University in 1996. After a two-year post-doc at the Hoover Institution, a thinktank on the Stanford campus, he became an assistant professor of managerial economics and decision sciences at the Kellogg School of Management, Northwestern University. He was promoted to associate professor in 2001 and received tenure at Kellogg in 2005. He continued to work at Kellogg until his departure to Zurich.
His research focuses on computational economics and finance. He applies numerical solution techniques to complex economic and financial models shedding light on relevant practical problems. He has published numerous research articles in international academic journals such as Econometrica, The Review of Economic Studies, The Journal of Finance, and The Review of Financial Studies, among others.
In developing new products, firms must balance entering into new domains (high uncertainty) with exploiting old domains (high competition). We leverage a unique database and rebuild the drug development pipelines of the Top 15 pharmaceutical companies between 1999-2016 in order to examine how firms select which projects to pursue and what impacts project success and failure. We find that firms select projects where they have prior experience, but that selection also depends on technological signals from rivals. Early-stage, uninformative technological signals increase the likelihood that the firm will diversify its search efforts to other domains, whereas late-stage, informative signals increase the likelihood of moving into the domain, but only if the rivals do not have a substantially large head-start. Moreover, conditional on selecting a project for further development, prior successful investments inform the firm as to what solutions work in a domain and increase the likelihood of future success. Prior failed investments inform as to what does not work and increase the likelihood of terminating future projects earlier.
Panos Markou is a Research Assistant at the Entrepreneurship at Cambridge Judge Business School. He is a strong believer in research that is grounded in practice and has the potential for large impact and relevance. Currently, he is working with BMW to analyse their portfolio of purchasing contracts and optimise their procurement and financial hedging strategy. During his doctoral studies, he worked with BMW to develop and implement a tool to evaluate commodity price indices for index-linked purchase contracts. Prior to this, Panos collaborated with Banco Santander in organising and hosting the 6th Annual Supply Chain Finance Symposium, and with Delta Air Lines in the TechOps Division. His research interests include the interface of finance, operations, and risk management; supply chain finance; empirical operations management.
The World Health Organization seeks effective ways to alert its member states about global pandemics. Motivated by this challenge, this study focuses on a public agency’s problem of designing warning policies to mitigate potential disasters that occur with advance notice. The agency privately receives early information about recurring harmful events and issues warnings to induce an uninformed party to take costly pre-emptive actions. The agency’s decision about whether to issue a warning critically depends on its credibility, which we define as the uninformed party’s belief regarding the accuracy of the agency’s information. This belief is updated over time by comparing the agency’s warnings with the actual incidence of harmful events. The sender, therefore, faces a trade-off between eliciting a proper response today and maintaining her credibility in order to elicit responses to future adverse events. The study formulates this problem as a dynamic Bayesian persuasion game, which is solved in closed form. Findings show that the agency must be sufficiently credible to elicit a mitigating action from the uninformed party for a given period. More importantly, the agency sometimes strategically misrepresents its advance information about a current threat in order to cultivate its future credibility. When its credibility is low (for example, below a threshold), the agency downplays the risk and actually downplays more as its credibility improves. By contrast, when its credibility is high (for example, above a second higher threshold), the agency sometimes exaggerates the threat. In this case, a less credible agency exaggerates more. Only when the agency’s credibility is moderate does it consistently send warning messages that fully disclose its private information about a potential disaster. These findings provide prescriptive guidelines for designing warning policies and suggest a plausible rationale for some of the false alarms or omissions observed in practice.
Francis de Véricourt is Professor of Management Science at ESMT European School of Management and Technology. From August 2010 until August 2013 he was on leave at INSEAD, where he was an Associate Professor of Technology and Operations Management and the Paul Dubrule Chaired Professor of Sustainable Development. He was the Associate Dean of Research at ESMT from 2007 to 2010. Before joining ESMT in 2007, Francis was an Associate Professor of Operations Management at Fuqua School of Business, Duke University. In 2000, he was a post-doctoral researcher at Massachusetts Institute of Technology (MIT). Francis received his PhD in Operations Research with Honours from Université Paris VI, France, in 2000. He holds an honours degree in Engineering in Applied Mathematics and Information Technology from Ecole Nationale Supérieure d’lnformatique et de Mathematiques Appliquées de Grenoble (ENSIMAG).
Francis’s general research interest is in the area of data-driven and managerial decision-making, with a current focus on healthcare, business sustainability, and service systems. He is the author of many research articles and has extensively published in leading academic journals, including Management Science, Operations Research, and American Economic Review. For his research, he has received a number of outstanding awards, including the 2011 MSOM best paper award of the Institute for Operations Research and the Management Sciences. He also holds editorial positions in flagship journals in operations research and management science.
Francis received numerous teaching awards for delivering classes to MBA and Executive MBA students at ESMT and INSEAD. He frequently teaches Executive Education Programs and is a regular speaker in academic and industry forums.
This seminar focuses on the study of a service setting where the provider may have advance information about customers’ future service needs and may initiate service for these customers proactively. Information about future customer service needs is becoming increasingly available due to better system integration coupled with advanced analytics and Big Data methods. To study this setting, the research combines (i) queueing theory, and in particular a diffusion approximation developed specifically for this problem, to quantify the impact of proactive service on customer delays with (ii) game theory to investigate the incentives of customers to agree to be served proactively. Findings show that proactive service reduces average delays, and we develop a closed-form approximation that shows that the benefit of proactive service is increasing concave in the proportion of customers who can be served proactively. Nevertheless, the study finds that in equilibrium, customers are less willing to agree to be served proactively compared to social optimum because of a positive externality leading to free riding behaviour, customers who agree to be served proactively reduce the waiting time for everyone, including those customers who do not have to suffer the inconvenience of being served proactively. The results suggest that proactive service may have a large operational benefit, but caution that it may fail to fulfil its potential due to customer self-interested behaviour.
Joint work with Kraig Delana and Nicos Savva.
Dr Tolga Tezcan is an Associate Professor of Management Science and Operations at London Business School. He teaches courses in Data Mining and Business Analytics. Prior to joining LBS, he was a faculty member at Simon School of Business at the University of Rochester between 2010 and 2015, where he was placed in the teaching honour roll in 2014, and at the University of Illinois at Urbana-Champaign between 2006 and 2010.
Tolga holds a PhD in Industrial and Systems Engineering and an MS in Mathematics from Georgia Tech, an MS in Industrial and Systems Engineering from Colorado State-Pueblo, and a BS in Industrial Engineering from Bilkent University, Turkey. Tolga’s research focuses on the robust management of service systems, such as customer service centres and healthcare systems, under uncertainty. His research has appeared in leading journals such as Management Science, Operations Research, Manufacturing & Service Operations Management, and Annals of Applied Probability. He has received the Career Award from National Science Foundation (NSF) USA in 2010.