Joseph Milner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Joseph Milner has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Opher Baron
Rotman School of Management, University of Toronto, Toronto, Ontario M5S 3E6, Canada
2 shared publications - Anton Ovchinnikov
Darden School of Business, University of Virginia, Charlottesville, Virginia, 39040, USA
1 shared publication - Gonzalo Romero
Rotman School of Management, University of Toronto, Toronto, Ontario M5S 3E6, Canada
1 shared publication - Hussein Naseraldin
Department of Industrial Engineering and Management, Ort Braude College, Karmiel, Israel
1 shared publication - Mahsa Hosseini
Odette School of Business, University of Windsor, Windsor, Ontario N9B 3P4, Canada
1 shared publication - Ningyuan Chen
Rotman School of Management, University of Toronto, Toronto, Ontario M5S 3E6, Canada
1 shared publication
Latest Publications
- 15 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Using Neural Networks to Guide Data-Driven Operational Decisions
We propose deep neural networks for data-driven stochastic optimization. Using historical data (covariates, decisions, costs), we propose to train a neural network to predict the objective value as a function of both the decision and covariate. After training, for a given covariate, this predicted objective is optimized over the decision variables using gradient-based methods with analytical gradients and Hessians. Performance is characterized by neural network generalization bounds. Comprehensive experiments on newsvendor, personalized assortment pricing, and call center staffing problems demonstrate our method’s strength over existing approaches such as conditional stochastic optimization and analytical approximations, especially when (i) the objective function is unknown, (ii) moderate to large data sets are available, or (iii) the problem structure resists simple parametric approximations. This paper was accepted by Chung Piaw Teo, optimization and decision analytics. Funding: The research of N. Chen is supported by the UTMM MARC Grant and the IMI Research Grant. The research of J. Milner is supported by NSERC [Grant 453954]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.04141 .
- 15 June 2026
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Management Science
Using Neural Networks to Guide Data-Driven Operational Decisions
We propose deep neural networks for data-driven stochastic optimization. Using historical data (covariates, decisions, costs), we propose to train a neural network to predict the objective value as a function of both the decision and covariate. After training, for a given covariate, this predicted objective is optimized over the decision variables using gradient-based methods with analytical gradients and Hessians. Performance is characterized by neural network generalization bounds. Comprehensive experiments on newsvendor, personalized assortment pricing, and call center staffing problems demonstrate our method’s strength over existing approaches such as conditional stochastic optimization and analytical approximations, especially when (i) the objective function is unknown, (ii) moderate to large data sets are available, or (iii) the problem structure resists simple parametric approximations. This paper was accepted by Chung Piaw Teo, optimization and decision analytics. Funding: The research of N. Chen is supported by the UTMM MARC Grant and the IMI Research Grant. The research of J. Milner is supported by NSERC [Grant 453954]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.04141 .
- 1 July 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (4) , 2010-2025
Dynamic Relocations in Car-Sharing Networks
In the paper “Dynamic Relocations in Car-Sharing Networks,” researchers have developed a new dynamic car relocation policy aimed at improving the efficiency of car-sharing networks. The policy addresses the challenge of unpredictable and uneven demand across the network by projecting the classical fluid model onto the simpler dimension of relocation decisions. This projection leads to a clearer understanding of the fluid model as a set of linear programs. The effectiveness of this new dynamic policy has been rigorously tested against traditional static policies through extensive simulations. The results are promising, showing that the dynamic policy significantly outperforms its static counterpart, reducing the optimality gap by more than 23% on average. This breakthrough offers substantial potential benefits for car-sharing operations, enhancing their ability to respond adaptively to changing demand and thus optimizing resource allocation and user satisfaction.
- 1 July 2025
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 73 (4) , 2010-2025
Dynamic Relocations in Car-Sharing Networks
In the paper “Dynamic Relocations in Car-Sharing Networks,” researchers have developed a new dynamic car relocation policy aimed at improving the efficiency of car-sharing networks. The policy addresses the challenge of unpredictable and uneven demand across the network by projecting the classical fluid model onto the simpler dimension of relocation decisions. This projection leads to a clearer understanding of the fluid model as a set of linear programs. The effectiveness of this new dynamic policy has been rigorously tested against traditional static policies through extensive simulations. The results are promising, showing that the dynamic policy significantly outperforms its static counterpart, reducing the optimality gap by more than 23% on average. This breakthrough offers substantial potential benefits for car-sharing operations, enhancing their ability to respond adaptively to changing demand and thus optimizing resource allocation and user satisfaction.
- 28 May 2019
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 67 (4) , 1184-1208
Pricing and Prioritizing Time-Sensitive Customers with Heterogeneous Demand Rates
Service firms such as amusement parks, ski resorts, and cultural institutions sell season passes and memberships that provide prioritized entry, while also offering nonprioritized pay-per-use acces... We consider the pricing/lead-time menu design problem for a monopoly service in which time-sensitive customers have demand on multiple occasions. Customers differ in their demand rates and marginal...
- 28 May 2019
- Published by Institute for Operations Research and the Management Sciences (INFORMS) in Operations Research
- Vol. 67 (4) , 1184-1208
Pricing and Prioritizing Time-Sensitive Customers with Heterogeneous Demand Rates
Service firms such as amusement parks, ski resorts, and cultural institutions sell season passes and memberships that provide prioritized entry, while also offering nonprioritized pay-per-use acces... We consider the pricing/lead-time menu design problem for a monopoly service in which time-sensitive customers have demand on multiple occasions. Customers differ in their demand rates and marginal...
- 1 August 2016
- Published by SAGE Publications in Production and Operations Management
- Vol. 25 (8) , 1344-1362
Inventory Rationing for Multiple Class Demand under Continuous Review
We consider how a firm should ration inventory to multiple classes in a stochastic demand environment with partial, class‐dependent backlogging where the firm incurs a fixed setup cost when ordering from its supplier. We present an infinite‐horizon, average cost criterion Markov decision problem formulation for the case with zero lead times. We provide an algorithm that determines the optimal rationing policy, and show how to find the optimal base‐stock reorder policy. Numerical studies indicate that the optimal policy is similar to that given by the equivalent deterministic problem and relies on tracking both the current inventory and the rate that backorder costs are accumulating. Our study of the case of non‐zero lead time shows that a heuristic combining the optimal, zero lead time policy with an allocation policy based on a single‐period profit management problem is effective.
- 1 August 2016
- Published by SAGE Publications in Production and Operations Management
- Vol. 25 (8) , 1344-1362
Inventory Rationing for Multiple Class Demand under Continuous Review
We consider how a firm should ration inventory to multiple classes in a stochastic demand environment with partial, class‐dependent backlogging where the firm incurs a fixed setup cost when ordering from its supplier. We present an infinite‐horizon, average cost criterion Markov decision problem formulation for the case with zero lead times. We provide an algorithm that determines the optimal rationing policy, and show how to find the optimal base‐stock reorder policy. Numerical studies indicate that the optimal policy is similar to that given by the equivalent deterministic problem and relies on tracking both the current inventory and the rate that backorder costs are accumulating. Our study of the case of non‐zero lead time shows that a heuristic combining the optimal, zero lead time policy with an allocation policy based on a single‐period profit management problem is effective.
- 1 January 2012
- Published by SAGE Publications in Production and Operations Management
- Vol. 21 (1) , 69-84
Revenue Management with End‐of‐Period Discounts in the Presence of Customer Learning
Consider a firm that sells identical products over a series of selling periods (e.g., weekly all‐inclusive vacations at the same resort). To stimulate demand and enhance revenue, in some periods, the firm may choose to offer a part of its available inventory at a discount. As customers learn to expect such discounts, a fraction may wait rather than purchase at a regular price. A problem the firm faces is how to incorporate this waiting and learning into its revenue management decisions. To address this problem we summarize two types of learning behaviors and propose a general model that allows for both stochastic consumer demand and stochastic waiting. For the case with two customer classes, we develop a novel solution approach to the resulting dynamic program. We then examine two simplified models, where either the demand or the waiting behavior are deterministic, and present the solution in a closed form. We extend the model to incorporate three customer classes and discuss the effects of overselling the capacity and bumping customers. Through numerical simulations we study the value of offering end‐of‐period deals optimally and analyze how this value changes under different consumer behavior and demand scenarios.
- 1 January 2012
- Published by SAGE Publications in Production and Operations Management
- Vol. 21 (1) , 69-84
Revenue Management with End‐of‐Period Discounts in the Presence of Customer Learning
Consider a firm that sells identical products over a series of selling periods (e.g., weekly all‐inclusive vacations at the same resort). To stimulate demand and enhance revenue, in some periods, the firm may choose to offer a part of its available inventory at a discount. As customers learn to expect such discounts, a fraction may wait rather than purchase at a regular price. A problem the firm faces is how to incorporate this waiting and learning into its revenue management decisions. To address this problem we summarize two types of learning behaviors and propose a general model that allows for both stochastic consumer demand and stochastic waiting. For the case with two customer classes, we develop a novel solution approach to the resulting dynamic program. We then examine two simplified models, where either the demand or the waiting behavior are deterministic, and present the solution in a closed form. We extend the model to incorporate three customer classes and discuss the effects of overselling the capacity and bumping customers. Through numerical simulations we study the value of offering end‐of‐period deals optimally and analyze how this value changes under different consumer behavior and demand scenarios.
- 15 November 2010
- Published by SAGE Publications in Production and Operations Management
- Vol. 20 (5) , 772-785
Facility Location: A Robust Optimization Approach
In this research, we apply robust optimization (RO) to the problem of locating facilities in a network facing uncertain demand over multiple periods. We consider a multi‐period fixed‐charge network location problem for which we find (1) the number of facilities, their location and capacities, (2) the production in each period, and (3) allocation of demand to facilities. Using the RO approach we formulate the problem to include alternate levels of uncertainty over the periods. We consider two models of demand uncertainty: demand within a bounded and symmetric multi‐dimensional box, and demand within a multi‐dimensional ellipsoid. We evaluate the potential benefits of applying the RO approach in our setting using an extensive numerical study. We show that the alternate models of uncertainty lead to very different solution network topologies, with the model with box uncertainty set opening fewer, larger facilities. Through sample path testing, we show that both the box and ellipsoidal uncertainty cases can provide small but significant improvements over the solution to the problem when demand is deterministic and set at its nominal value. For changes in several environmental parameters, we explore the effects on the solution performance.
- 15 November 2010
- Published by SAGE Publications in Production and Operations Management
- Vol. 20 (5) , 772-785
Facility Location: A Robust Optimization Approach
In this research, we apply robust optimization (RO) to the problem of locating facilities in a network facing uncertain demand over multiple periods. We consider a multi‐period fixed‐charge network location problem for which we find (1) the number of facilities, their location and capacities, (2) the production in each period, and (3) allocation of demand to facilities. Using the RO approach we formulate the problem to include alternate levels of uncertainty over the periods. We consider two models of demand uncertainty: demand within a bounded and symmetric multi‐dimensional box, and demand within a multi‐dimensional ellipsoid. We evaluate the potential benefits of applying the RO approach in our setting using an extensive numerical study. We show that the alternate models of uncertainty lead to very different solution network topologies, with the model with box uncertainty set opening fewer, larger facilities. Through sample path testing, we show that both the box and ellipsoidal uncertainty cases can provide small but significant improvements over the solution to the problem when demand is deterministic and set at its nominal value. For changes in several environmental parameters, we explore the effects on the solution performance.