Diwakar Gupta
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Diwakar Gupta has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
George Shanthikumar
University of California
2 shared publications- Haresh Gurnani
Department of Management, University of Miami, Coral Gables, Florida 33124
2 shared publications Ajay K. Israni
University of Minnesota
1 shared publicationAmiram Gafni
Centre for Health Economics and Policy Analysis, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada
1 shared publicationBertram L. Kasiske
Department of Medicine, Hennepin County Medical Center; Scientific Registry of Transplant Recipients, Minneapolis Medical Research Foundation
1 shared publicationDon Shilton
Population Health Research Institute, Department of Medicine, McMaster University, Hamilton, Ontario, Canada
1 shared publication
Latest Publications
- 17 February 2015
- Published by Springer Nature in Health Care Management Science
- Vol. 19 (3) , 261-278
Improving operating room schedules
Operating rooms (ORs) in US hospitals are costly to staff, generate about 70 % of a hospital’s revenues, and operate at a staffed-capacity utilization of 60-70 %. Many hospitals allocate blocks of OR time to individual or groups of surgeons as guaranteed allocation, who book surgeries one at a time in their blocks. The booking procedure frequently results in unused time between surgeries. Realizing that this presents an opportunity to improve OR utilization, hospitals manually reschedule surgery start times one or two days before each day of surgical operations. The purpose of rescheduling is to decrease OR staffing costs, which are determined by the number of concurrently staffed ORs. We formulate the rescheduling problem as a variant of the bin-packing problem with interrelated items, which are the surgeries performed by the same surgeon. We develop a lower bound (LB) construction algorithm and prove that the LB is at least (2/3) of the optimal staffing cost. A key feature of our approach is that we allow hospitals to have two shift lengths. Our analytical results form the basis of a branch-and-bound algorithm, which we test on data obtained from three hospitals. Experiments show that rescheduling saves significant staffing costs.
- 17 February 2015
- Published by Springer Nature in Health Care Management Science
- Vol. 19 (3) , 261-278
Improving operating room schedules
Operating rooms (ORs) in US hospitals are costly to staff, generate about 70 % of a hospital’s revenues, and operate at a staffed-capacity utilization of 60-70 %. Many hospitals allocate blocks of OR time to individual or groups of surgeons as guaranteed allocation, who book surgeries one at a time in their blocks. The booking procedure frequently results in unused time between surgeries. Realizing that this presents an opportunity to improve OR utilization, hospitals manually reschedule surgery start times one or two days before each day of surgical operations. The purpose of rescheduling is to decrease OR staffing costs, which are determined by the number of concurrently staffed ORs. We formulate the rescheduling problem as a variant of the bin-packing problem with interrelated items, which are the surgeries performed by the same surgeon. We develop a lower bound (LB) construction algorithm and prove that the LB is at least (2/3) of the optimal staffing cost. A key feature of our approach is that we allow hospitals to have two shift lengths. Our analytical results form the basis of a branch-and-bound algorithm, which we test on data obtained from three hospitals. Experiments show that rescheduling saves significant staffing costs.
- 30 August 2014
- Published by Springer Nature in Health Care Management Science
- Vol. 18 (1) , 35-57
Accept/decline decision module for the liver simulated allocation model
Simulated allocation models (SAMs) are used to evaluate organ allocation policies. An important component of SAMs is a module that decides whether each potential recipient will accept an offered organ. The objective of this study was to develop and test accept-or-decline classifiers based on several machine-learning methods in an effort to improve the SAM for liver allocation. Feature selection and imbalance correction methods were tested and best approaches identified for application to organ transplant data. Then, we used 2011 liver match-run data to compare classifiers based on logistic regression, support vector machines, boosting, classification and regression trees, and Random Forests. Finally, because the accept-or-decline module will be embedded in a simulation model, we also developed an evaluation tool for comparing performance of predictors, which we call sample-path accuracy. The Random Forest method resulted in the smallest overall error rate, and boosting techniques had greater accuracy when both sensitivity and specificity were simultaneously considered important. Our comparisons show that no method dominates all others on all performance measures of interest. A logistic regression-based classifier is easy to implement and allows for pinpointing the contribution of each feature toward the probability of acceptance. Other methods we tested did not have a similar interpretation. The Scientific Registry of Transplant Recipients decided to use the logistic regression-based accept-decline decision module in the next generation of liver SAM.
- 30 August 2014
- Published by Springer Nature in Health Care Management Science
- Vol. 18 (1) , 35-57
Accept/decline decision module for the liver simulated allocation model
Simulated allocation models (SAMs) are used to evaluate organ allocation policies. An important component of SAMs is a module that decides whether each potential recipient will accept an offered organ. The objective of this study was to develop and test accept-or-decline classifiers based on several machine-learning methods in an effort to improve the SAM for liver allocation. Feature selection and imbalance correction methods were tested and best approaches identified for application to organ transplant data. Then, we used 2011 liver match-run data to compare classifiers based on logistic regression, support vector machines, boosting, classification and regression trees, and Random Forests. Finally, because the accept-or-decline module will be embedded in a simulation model, we also developed an evaluation tool for comparing performance of predictors, which we call sample-path accuracy. The Random Forest method resulted in the smallest overall error rate, and boosting techniques had greater accuracy when both sensitivity and specificity were simultaneously considered important. Our comparisons show that no method dominates all others on all performance measures of interest. A logistic regression-based classifier is easy to implement and allows for pinpointing the contribution of each feature toward the probability of acceptance. Other methods we tested did not have a similar interpretation. The Scientific Registry of Transplant Recipients decided to use the logistic regression-based accept-decline decision module in the next generation of liver SAM.
- 1 July 2014
- Published by SAGE Publications in Production and Operations Management
- Vol. 23 (7) , 1198-1211
Managing Disruptions in Decentralized Supply Chains with Endogenous Supply Process Reliability
Supply disruptions are all too common in supply chains. To mitigate delivery risk, buyers may either source from multiple suppliers or offer incentives to their preferred supplier to improve its process reliability. These incentives can be either direct (investment subsidy) or indirect (inflated order quantity). In this study, we present a series of models to highlight buyers’ and suppliers’ optimal parameter choices. Our base‐case model has deterministic buyer demand and two possibilities for the supplier yield outcomes: all‐or‐nothing supply or partial disruption. For the all‐or‐nothing model, we show that the buyer prefers to only use the subsidy option, which obviates the need to inflate order quantity. However, in the partial disruption model, both incentives—subsidy and order inflation—may be used at the same time. Although single sourcing provides greater indirect incentive to the selected supplier because that avoids order splitting, we show that the buyer may prefer the diversification strategy under certain circumstances. We also quantify the amount by which the wholesale price needs to be discounted (if at all) to ensure that dual sourcing strategy dominates sole sourcing. Finally, we extend the model to the case of stochastic demand. Structural properties of ordering/subsidy decisions are derived for the all‐or‐nothing model, and in contrast to the deterministic demand case, we establish that the buyer may increase use of subsidy and order quantity at the same time.
- 1 July 2014
- Published by SAGE Publications in Production and Operations Management
- Vol. 23 (7) , 1198-1211
Managing Disruptions in Decentralized Supply Chains with Endogenous Supply Process Reliability
Supply disruptions are all too common in supply chains. To mitigate delivery risk, buyers may either source from multiple suppliers or offer incentives to their preferred supplier to improve its process reliability. These incentives can be either direct (investment subsidy) or indirect (inflated order quantity). In this study, we present a series of models to highlight buyers’ and suppliers’ optimal parameter choices. Our base‐case model has deterministic buyer demand and two possibilities for the supplier yield outcomes: all‐or‐nothing supply or partial disruption. For the all‐or‐nothing model, we show that the buyer prefers to only use the subsidy option, which obviates the need to inflate order quantity. However, in the partial disruption model, both incentives—subsidy and order inflation—may be used at the same time. Although single sourcing provides greater indirect incentive to the selected supplier because that avoids order splitting, we show that the buyer may prefer the diversification strategy under certain circumstances. We also quantify the amount by which the wholesale price needs to be discounted (if at all) to ensure that dual sourcing strategy dominates sole sourcing. Finally, we extend the model to the case of stochastic demand. Structural properties of ordering/subsidy decisions are derived for the all‐or‐nothing model, and in contrast to the deterministic demand case, we establish that the buyer may increase use of subsidy and order quantity at the same time.
- 27 January 2014
- Published by Taylor & Francis in IIE Transactions
- Vol. 46 (11) , 1132-1146
The extraboard operator scheduling and work assignment problem
An instance of the operational fixed job scheduling problem arises when open work caused by unplanned events such as bus breakdowns, inclement weather, and driver (operator) absenteeism need to be covered by reserve (extraboard) drivers. Each work-piece, which is referred to as a job, requires one operator who must work continuously between specified start and end times to complete the job. Each extraboard operator may be assigned up to w hours of work, which may not to be continuous so long as the total work time is within a s-hour time window of that operator’s shift start time. Parameters w and s are called allowable work-time and spread-time, respectively. The objective is to choose operators’ shift start times and work assignments, while honoring work-time and spread-time constraints, such that the amount of work covered as part of regular duties is maximized. This paper argues that the extraboard operator scheduling problem is NP-hard and three heuristic approaches are presented for the solution of such problems. These include a decomposition-based algorithm whose worst-case performance ratio is proved to lie in [1 − 1/e, 19/27], where e ≈ 2:718 is the base of the natural logarithm. Numerical experiments are presented that use data from a large transit agency, which show that the average performance of the decomposition algorithm is good when applied to real-world data.
- 27 January 2014
- Published by Taylor & Francis in IIE Transactions
- Vol. 46 (11) , 1132-1146
The extraboard operator scheduling and work assignment problem
An instance of the operational fixed job scheduling problem arises when open work caused by unplanned events such as bus breakdowns, inclement weather, and driver (operator) absenteeism need to be covered by reserve (extraboard) drivers. Each work-piece, which is referred to as a job, requires one operator who must work continuously between specified start and end times to complete the job. Each extraboard operator may be assigned up to w hours of work, which may not to be continuous so long as the total work time is within a s-hour time window of that operator’s shift start time. Parameters w and s are called allowable work-time and spread-time, respectively. The objective is to choose operators’ shift start times and work assignments, while honoring work-time and spread-time constraints, such that the amount of work covered as part of regular duties is maximized. This paper argues that the extraboard operator scheduling problem is NP-hard and three heuristic approaches are presented for the solution of such problems. These include a decomposition-based algorithm whose worst-case performance ratio is proved to lie in [1 − 1/e, 19/27], where e ≈ 2:718 is the base of the natural logarithm. Numerical experiments are presented that use data from a large transit agency, which show that the average performance of the decomposition algorithm is good when applied to real-world data.
- 1 August 2013
- Published by Taylor & Francis in IIE Transactions
- Vol. 45 (8) , 811-825
Fast-ship commitment contracts in retail supply chains
This article analyzes three types of supply contracts between a supplier and a retailer when both agree as follows—if a customer experiences a stockout, then the purchased item can be shipped to the customer on an expedited basis at no extra cost. This practice is referred to as the fast-ship option in this article. In the first contract (Structure A), the supplier specifies a total supply commitment and allows the retailer to choose its split between the initial order and the amount left to satisfy fast-ship orders. In the other two contracts (Structures B and C), the supplier agrees to fully supply the retailer’s initial order but places a restriction on the quantity available for fast-ship commitment. The difference between the second and third contracts is that in contract Structure B the supplier moves first, whereas in contract Structure C the supplier determines its commitment after observing the retailer’s order. The supplier’s and the retailer’s optimal decisions and preferences are characterized. The question of how the supplier and the retailer may resolve their conflict regarding the preferred contract type is addressed. [Supplementary materials are available for this article. Go to the publisher’s online edition of IIE Transactions for proofs.]
- 1 August 2013
- Published by Taylor & Francis in IIE Transactions
- Vol. 45 (8) , 811-825
Fast-ship commitment contracts in retail supply chains
This article analyzes three types of supply contracts between a supplier and a retailer when both agree as follows—if a customer experiences a stockout, then the purchased item can be shipped to the customer on an expedited basis at no extra cost. This practice is referred to as the fast-ship option in this article. In the first contract (Structure A), the supplier specifies a total supply commitment and allows the retailer to choose its split between the initial order and the amount left to satisfy fast-ship orders. In the other two contracts (Structures B and C), the supplier agrees to fully supply the retailer’s initial order but places a restriction on the quantity available for fast-ship commitment. The difference between the second and third contracts is that in contract Structure B the supplier moves first, whereas in contract Structure C the supplier determines its commitment after observing the retailer’s order. The supplier’s and the retailer’s optimal decisions and preferences are characterized. The question of how the supplier and the retailer may resolve their conflict regarding the preferred contract type is addressed. [Supplementary materials are available for this article. Go to the publisher’s online edition of IIE Transactions for proofs.]
- 1 January 2011
- Published by SAGE Publications in Production and Operations Management
- Vol. 20 (1) , 152-162
A Note on Air‐Cargo Capacity Contracts
Carriers (airlines) use medium‐term contracts to allot bulk cargo capacity to forwarders who deliver consolidated loads for each flight in the contractual period (season). Carriers also sell capacity to direct‐ship customers on each flight. We study capacity contracts between a carrier and a forwarder when certain parameters such as the forwarder's demand, operating cost to the carrier, margin, and reservation profit are its private information. We propose contracts in which the forwarder pays a lump sum in exchange for a guaranteed capacity allotment and receives a refund for each unit of unused capacity according to a pre‐announced refund rate. We obtain an upper bound on the informational rent paid by the carrier for a menu of arbitrary allotments and identify conditions under which it can eliminate the informational rent and induce the forwarder to choose the overall optimal capacity allotment (i.e., one that maximizes the combined profits of the carrier and the forwarder).
- 1 January 2011
- Published by SAGE Publications in Production and Operations Management
- Vol. 20 (1) , 152-162
A Note on Air‐Cargo Capacity Contracts
Carriers (airlines) use medium‐term contracts to allot bulk cargo capacity to forwarders who deliver consolidated loads for each flight in the contractual period (season). Carriers also sell capacity to direct‐ship customers on each flight. We study capacity contracts between a carrier and a forwarder when certain parameters such as the forwarder's demand, operating cost to the carrier, margin, and reservation profit are its private information. We propose contracts in which the forwarder pays a lump sum in exchange for a guaranteed capacity allotment and receives a refund for each unit of unused capacity according to a pre‐announced refund rate. We obtain an upper bound on the informational rent paid by the carrier for a menu of arbitrary allotments and identify conditions under which it can eliminate the informational rent and induce the forwarder to choose the overall optimal capacity allotment (i.e., one that maximizes the combined profits of the carrier and the forwarder).
- 31 December 2009
- Published by Elsevier in Health Policy
- Vol. 93 (2-3) , 188-200
On evaluating the impact of flexibility enhancing strategies on the performance of nurse schedules
- 31 December 2009
- Published by Elsevier in Health Policy
- Vol. 93 (2-3) , 188-200
On evaluating the impact of flexibility enhancing strategies on the performance of nurse schedules
- 21 June 2008
- Published by Springer Nature in Flexible Services and Manufacturing Journal
- Vol. 20 (1-2) , 11-12
In this issue
- 21 June 2008
- Published by Springer Nature in Flexible Services and Manufacturing Journal
- Vol. 20 (1-2) , 11-12
In this issue