ALBERT

Description: 〈p〉Publication date: Available online 8 July 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): K.T. Huynh〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We are interested in the stochastic modeling of a condition-based maintained system subject to continuous deterioration and maintenance actions such as inspection, partial repair and replacement. The partial repair is assumed dependent on the past in the sense that it cannot bring the system back into a deterioration state better than the one reached at the last repair. Such a past-dependency can affect (〈em〉i〈/em〉) the selection of a type of maintenance actions, (〈em〉ii〈/em〉) the maintenance duration, (〈em〉iii〈/em〉) the deterioration level after a maintenance, and (〈em〉iv〈/em〉) the restarting system deterioration behavior. In this paper, all these effects are jointly considered in an unifying condition-based maintenance model on the basis of restarting deterioration states randomly sampled from a probability distribution truncated by the deterioration levels just before a current repair and just after the last repair/replacement. Using results from the semi-regenerative theory, the long-run maintenance cost rate is analytically derived. Numerous sensitivity studies illustrate the impacts of past-dependent partial repairs on the economic performance of the considered condition-based maintained system.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 5 July 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Shichen Zhang, Jianxiong Zhang〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Nowadays, some suppliers are looking for offline expansion in addition to their preexisting online channels relying on e-tailers. This study focuses on the e-tailer’s demand information sharing strategy with the supplier who may build upon brick-and-mortar stores. Both prevailing agreements between the supplier and the e-tailer are investigated: agency selling and reselling. The equilibrium results are quite different under these two agreements. Specifically, when the supplier’s offline entry cost is very small or large, the e-tailer shares information under agency selling while keeps information private under reselling. When the entry cost is intermediate, channel substitution rate is large and information uncertainty is small, the e-tailer withholds the demand information under agency selling while shares information under reselling to deter the supplier from entering an offline channel. Furthermore, two extensions about consumer behavior in multichannel selection are discussed: showrooming and webrooming. With showrooming or webrooming, the e-tailer’s information sharing decisions qualitatively hold, while with showrooming the drive factor behind may change; that is, withholding information under agency selling and sharing information under reselling may also serve as measures to encourage supplier offline entry when the effect of showrooming is strong.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 2 July 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Kieran Conboy, Patrick Mikalef, Denis Dennehy, John Krogstie〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉While the topic of analytics is rapidly growing in popularity across various domains, there is still a relatively low amount of empirical work in the field of operations research (OR). While studies of various technical and business aspects of analytics are emerging in OR, little has been done to address how the OR community can leverage business analytics in dynamic and uncertain environments – the very place where OR is supposed to play a key role. To address this gap, this study draws on the dynamic capabilities view of the firm and builds on eight selected case studies of operations research activity in large organisations, each of which have invested significantly in analytics technology and implementation. The study identifies fourteen analytics-enabled micro-foundations of dynamic capabilities, essentially highlighting how organisations can use analytics to manage and enhance their OR activities in dynamic and uncertain environments. This study also identifies six key cross-cutting propositions emerging from the data and develops a roadmap for future OR researchers to address these issues and improve the use and value of analytics as enablers of organisational dynamic capabilities.〈/p〉〈/div〉

Description: 〈p〉Publication date: 1 December 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research, Volume 279, Issue 2〈/p〉 〈p〉Author(s): 〈/p〉

Description: 〈p〉Publication date: Available online 29 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Claudio Arbib, Mustafa Ç. Pınar, Matteo Tonelli〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Consider a three-level non-capacitated location/pricing problem: a firm first decides which facilities to open, out of a finite set of candidate sites, and sets service prices with the aim of revenue maximization; then a second firm makes the same decisions after checking competing offers; finally, customers make individual decisions trying to minimize costs that include both purchase and transportation. A restricted two-level problem can be defined to model an optimal reaction of the second firm to known decision of the first.〈/p〉 〈p〉For non-metric costs, the two-level problem corresponds to 〈span〉Envy-free Pricing〈/span〉 or to a special 〈span〉Network Pricing〈/span〉 problem, and is 〈math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si65.svg"〉〈mi mathvariant="bold-script"〉APX〈/mi〉〈/math〉-complete even if facilities can be opened at no fixed cost. Our focus is on the metric 1-dimensional case, a model where customers are distributed on a main communication road and transportation cost is proportional to distance. We describe polynomial-time algorithms that solve two- and three-level problems with opening costs and single 1〈sup〉〈em〉st〈/em〉〈/sup〉 level facility. Quite surprisingly, however, even the two-level problem with no opening costs becomes 〈math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si66.svg"〉〈mi mathvariant="bold-script"〉NP〈/mi〉〈/math〉-hard when two 1〈sup〉〈em〉st〈/em〉〈/sup〉 level facilities are considered.〈/p〉 〈/div〉

Description: 〈p〉Publication date: Available online 29 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Zachary Feinstein〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We construct a continuous time model for price-mediated contagion precipitated by a common exogenous stress to the banking book of all firms in the financial system. In this setting, firms are constrained so as to satisfy a risk-weight based capital ratio requirement. We use this model to find analytical bounds on the risk-weights for assets as a function of the market liquidity. Under these appropriate risk-weights, we find existence and uniqueness for the joint system of firm behavior and the asset prices. We further consider an analytical bound on the firm liquidations, which allows us to construct exact formulas for stress testing the financial system with deterministic or random stresses. Numerical case studies are provided to demonstrate various implications of this model and analytical bounds.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 29 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Bo Jin〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In a recent paper, Galle, Barnhart, and Jaillet [Galle, V., Barnhart, C., & Jaillet, P. (2018). A new binary formulation of the restricted container relocation problem based on a binary encoding of configurations. 〈em〉European Journal of Operational Research, 267〈/em〉(2), 467–477] introduced a new variant of the container relocation problem (CRP), named the relaxed restricted CRP, where every container can be relocated at most once for retrieving each target container. The authors also proposed a binary integer programming model for formulating the relaxed restricted CRP. In this paper, it is first shown that the proposed model contains two deficiencies in formulating the “last in, first out” (LIFO) policy. These deficiencies will cause the solutions obtained by the model to correspond to infeasible configurations or infeasible relocation sequences. Then, the LIFO policy is analyzed in detail and reformulated as linear constraints correctly. Lastly, the corrected integer programming formulation is presented. Computational experiments show that the corrected model dramatically reduces complexity and improves performance.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 28 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Mike G. Tsionas〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Mitropoulos et al. (Mitropoulos P., M. A. Talias, and I. Mitropoulos, 2015, Combining stochastic DEA with Bayesian analysis to obtain statistical properties of the efficiency scores: An application to Greek public hospitals. European Journal of Operational Research 243, 302-311) suggested the use of a Bayesian approach in Data Envelopment Analysis (DEA) which can be used to obtain posterior distributions of efficiency scores. In this paper, we avoid their assumption that alternative data sets are simulated from the predictive distribution obtained from their simple data generating process of a normal distribution for the data. The new approach has two significant advantages. First, the posterior proposed in this paper is coherent or principled in the sense that it is consistent with the DEA formulation. Second, and perhaps surprisingly, it is not necessary to solve linear programming problems for each observation in the sample. Bayesian inference is organized around Markov Chain Monte Carlo techniques that can be implemented quite easily. We conduct extensive Monte Carlo experiments to investigate the finite-sample properties of the new approach. We also provide an application to a large U.S banking data set. The sample is an unbalanced panel of US banks with 2,397 bank–year observations for 285 banks. The main purpose of the analysis is to compare distributions of efficiency scores. Relative to DEA, Bayes DEA provides different efficiency scores and their sample distribution has significantly less probability concentration around unity. The comparison with bootstrap-DEA shows that results from Bayes DEA are in broad agreement.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 21 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Zelin Zhang, Minghui Ma, Peter T.L. Popkowski Leszczyc, Hejun Zhuang〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉This research proposes an analytical model of the joint optimization of coupon face value and duration together with the product price, and determines the impact of coupon design on consumers’ redemption behavior. A model of rational forward-looking consumers’ redemption behavior is derived that incorporates forgetting (to redeem) and stochastic redemption costs.〈/p〉 〈p〉Results show that when product price is exogenous, long-duration coupons may result in increased seller profits and always increase consumer surplus. Moreover, a one-period coupon is never optimal when (1) the difference in valuations for high-value (loyal) and low-value (non-loyal) consumers or (2) the coupon face value is larger than the redemption costs of high-value consumers. Long-duration coupons tend to be optimal when the level of recall of high-value consumers is sufficiently low, which reduces redemption by high-value consumers.〈/p〉 〈p〉Coupon duration together with face value plays an important role in coupons’ ability to price discriminate between different consumer segments and to avoid head-on competition with other sellers. Results can replicate empirically observed redemptions patterns, which has important implications for the strategic targeting of coupons to different consumer segments.〈/p〉 〈p〉A coupon may result in an increase or decrease in price. When the difference in valuation between high-value and low-value consumers is high (relative to the redemption costs), a seller can either reduce price and lower face value or increase coupon duration for the purpose of avoiding redemption by high-value consumers.〈/p〉 〈/div〉

Description: 〈p〉Publication date: Available online 21 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Sturla F. Kvamsdal, José M. Maroto, Manuel Morán, Leif K. Sandal〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Seasonality or periodicity–biological, environmental, or economic–are fundamental properties of most marine fisheries. We propose a generic infinite horizon discrete time fisheries management model by modifying existing models and frameworks of analysis to reflect seasonal variation or more general multiperiodicity in parameters, variables, or functional forms. Our model captures such variations via repeated cycles of multiple intervals with differing characteristics. The framework offers a simple and intuitive set up of arbitrary periodicity and seasonality in any feature, which significantly increases model realism. Further, it distances itself from continuous time modeling approaches where uniqueness and solvability of periodic models generally are difficult to assert. In our setting, the governing equations for the time-dependent value function of the management optimization problem are shown to be equivalent to a high-dimensional contraction and hence ensure uniqueness and a feasible solution algorithm. We illustrate our approach using a simple example to demonstrate that accounting for seasonality in fisheries management can improve outcomes considerably. Our framework also provides for analysis of seasonal regulatory measures. Ultimately, our approach applies to renewable resource management more generally and to many infinite-horizon, discrete time optimization problems with periodic features.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 20 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Chandra Ade Irawan, Majid Eskandarpour, Djamila Ouelhadj, Dylan Jones〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Scheduling maintenance routing for an offshore wind farm is a challenging and complex task. The problem is to find the best routes for the Crew Transfer Vessels to maintain the turbines in order to minimise the total cost. This paper primarily proposes an efficient solution method to solve the deterministic maintenance routing problem in an offshore wind farm. The proposed solution method is based on the Large Neighbourhood Search metaheuristic. The efficiency of the proposed metaheuristic is validated against state of the art algorithms. The results obtained from the computational experiments validate the effectiveness of the proposed method. In addition, as the maintenance activities are affected by uncertain conditions, a simulation-based optimisation algorithm is developed to tackle these uncertainties. This algorithm benefits from the fast computational time and solution quality of the proposed metaheuristic, combined with Monte Carlo simulation. The uncertain factors considered include the travel time for a vessel to visit turbines, the required time to maintain a turbine, and the transfer time for technicians and equipment to a turbine. Moreover, the proposed simulation-based optimisation algorithm is devised to tackle unpredictable broken-down turbines. The performance of this algorithm is evaluated using a case study based on a reference wind farm scenario developed in the EU FP7 LEANWIND project.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 20 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Simon J. Höller, Raik Özsen, Ulrich W. Thonemann〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Order expediting is an inventory control practice that allows companies to reduce inventory while maintaining service levels. However, expediting outstanding orders is costly and a trade-off must be made between expediting orders and holding inventory. We model the problem as a periodic-review inventory system with the option to move outstanding units forward in the replenishment pipeline. The objective is to minimize the sum of expected inventory holding and expediting costs per period subject to a minimum expected service level constraint. We consider a generalized base-stock policy where outstanding units are expedited when the inventory level drops below a certain threshold. We develop structural properties and present an efficient procedure to determine the optimal policy parameters. In a numerical study with real-world data, we show that our expediting policy offers substantial savings compared to other policies. We also provide managerial insights by numerically analyzing how the model parameters impact the savings.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 20 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): M. Fampa, D. Lubke, F. Wang, H. Wolkowicz〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We consider a parametric convex quadratic programming (CQP) relaxation for the quadratic knapsack problem (QKP). This relaxation maintains partial quadratic information from the original QKP by perturbing the objective function to obtain a concave quadratic term. The nonconcave part generated by the perturbation is then linearized by a standard approach that lifts the problem to matrix space. We present a primal-dual interior point method to optimize the perturbation of the quadratic function, in a search for the tightest upper bound for the QKP. We prove that the same perturbation approach, when applied in the context of semidefinite programming (SDP) relaxations of the QKP, cannot improve the upper bound given by the corresponding linear SDP relaxation. The result also applies to more general integer quadratic problems. Finally, we propose new valid inequalities on the lifted matrix variable, derived from cover and knapsack inequalities for the QKP, and present separation problems to generate cuts for the current solution of the CQP relaxation. Our best bounds are obtained alternating between optimizing the parametric quadratic relaxation over the perturbation and applying cutting planes generated by the valid inequalities proposed.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 20 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Yusuke Zennyo〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper examines strategic contracting between a monopoly platform and suppliers that sell their goods through the platform. I consider two competing suppliers: a high-volume supplier with the larger potential demand and a low-volume supplier with the smaller one. Each supplier chooses one of two contracts: wholesale or agency. The platform has to strategically determine the royalty rate for the agency contract by taking into account which contracts the suppliers will choose. I show that the platform offers a low (high) royalty rate to induce the suppliers to adopt the agency (wholesale) contract when product substitutability is low (high) enough. More interestingly, when the degree of substitution is at an intermediate level, asymmetric contracting, in which only the low-volume supplier adopts the agency contract, can arise in equilibrium. This result is related to the fact that many long-tail and niche products with lower potential market sizes are traded on platform-based marketplaces, such as Amazon Marketplace and Walmart Marketplace.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 31 July 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Shu Guo, Tsan-Ming Choi, Bin Shen〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Motivated by the observed industrial issues, we analytically develop a fashion supply chain consisting of one manufacturer and two competing retailers and investigate how retail competition and consumer returns affect green product development in fashion apparel. In the basic model, that is, the pure “product greenness level” game, we find that the optimal greenness level of the fashion product decreases along with the level of market competition. This finding implies that a more competitive market leads to a lower optimal greenness level. We also identify that when the consumer return rate increases, the optimal product greenness level is substantially reduced. In the extended model with joint decisions on greenness and pricing, we find that the optimal product greenness level for the whole channel is always higher in the scenario when both retailers charge a higher retail price than in the case with a lower retail price. As such, the underdevelopment of green fashion products is a result of fashion industry features, such as an extremely competitive environment for green product development, relatively low retail prices for fashion products, and high consumer return rates. Therefore, fashion companies should join a co-opetition game for the green product market and simultaneously enhance their efficiency in managing consumer returns. To support our analytical findings, we conduct extensive industrial interviews with various representative companies. Based on this multi-methodological approach (MMA), this paper generates practice-relevant managerial insights that not only contribute to the literature, but also act as valuable references for industrialists.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 31 July 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Daqiang Chen, Joshua Ignatius, Danzhi Sun, Mark Goh, Shalei Zhan〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper addresses the problem of the firms operating on cross-border or inter-regional platforms that are subject to the enforcement of each local government's carbon emissions regulatory policy, thus causing an imbalance in the sharing of the burden of the greening of the total supply chain. We introduce the concept of equity as the incentive mechanism to coordinate this green supply chain which is a function of the carbon emission permits and the revenue generated by the firms. Due to the complexity and imbalance in the original incentive mechanism to this problem, we provide a new equivalent supply chain network equilibrium model under elastic demand based on user equilibrium theory. We state the user equilibrium conditions and provide the equivalent formulation. We show the trade-offs under various carbon emissions regulatory policies. A product with higher price elasticity and carbon emission intensity not only hampers the firm from gaining a higher revenue, but it also reduces the equity of the system under an invariant emission regulatory policy.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 28 September 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Noemie Balouka, Izack Cohen〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper suggests a robust optimization approach for the multi-mode resource-constrained project scheduling problem with uncertain activity durations. The objective is to minimize the worst-case project duration by deciding on activity modes, resource allocations and a schedule baseline. The problem is solved by a Benders decomposition approach with specialized cuts. We consider polyhedral uncertainty sets in which the level of conservatism can be adjusted. Using a computational study in which various problem instances are explored under varying levels of uncertainty, conservatism and several types of duration distributions, we provide insights about the price of robustness and the performance of the approach. The hope is that these insights can guide future multi-mode project scheduling implementations when there is partial information about the distribution of activity durations.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 13 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Gilberto Montibeller, Pratik Patel, Victor J. del Rio Vilas〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Multi-criteria assessments are increasingly being employed in the prioritisation of health threats, supporting decision processes related to health risk management. The use of multi-criteria analysis in this context is welcome, as it facilitates the consideration of multiple impacts of health threats, it can encompass the use of expert judgment to complement and amalgamate the evidence available, and it permits the modelling of policy makers’ priorities. However, these assessments often lack a clear multi-criteria conceptual framework, in terms of both axiomatic rigour and adequate procedures for preference modelling. Such assessments are 〈em〉ad hoc〈/em〉 from a multi-criteria decision analysis perspective, despite the strong health expertise used in constructing these models. In this paper we critically examine some key assumptions and modelling choices made in these assessments, comparing them with the best practices of multi-attribute value analysis. Furthermore, we suggest a set of guidelines on how simulation studies might be employed to assess the impact of these modelling choices. We apply these guidelines to two relevant studies available in the health threat prioritisation domain. We identify severe variability in our simulations due to poor modelling choices, which could cause changes in the ranking of threats being assessed and thus lead to alternative policy recommendations than those suggested in their reports. Our results confirm the importance of carefully designing multi-criteria evaluation models for the prioritisation of health threats.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 13 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Antoine Lesage-Landry, Joshua A. Taylor〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We construct a model for transmission planning with both alternating and direct current lines, the latter of which can be interfaced via either line-commutated converters or voltage-source converters. The transmission expansion problem is nonlinear and nonconvex. Thus, nonlinear solvers cannot guarantee their convergence to the global optimum of the problem. We use relaxations and approximations to formulate a mixed-integer second-order cone transmission expansion model, which can be solved to optimality by current industrial solvers. We base our formulation on the branch flow relaxation. We include losses and reactive power placement, and consider direct current lines connected by both line-commutated converters and voltage-sourced converters. We show that our approach lowers the expansion cost on 6-bus and 24-bus system examples. We evaluate the feasibility of our formulation using a semidefinite relaxation of optimal power flow and find that the resulting plan admits feasible or close to feasible power flows.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 12 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Valentin Zelenyuk〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The main goal of this paper is to explore the possible solutions to a ‘big data’ problem related to the very large dimensions of input–output data. In particular, we focus on the cases of severe ‘curse of dimensionality’ problem that require dimension-reduction prior to using Data Envelopment Analysis. To achieve this goal, we have presented some theoretical grounds and performed a new to the literature simulation study where we explored the price-based aggregation as a solution to address the problem of very large dimensions.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 12 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): M. Densing〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A stored commodity is sold under a capacity constraint depending on an exogenous random market price, and, alternatively, a service contract can be provided in which the selling amount must be held constant over time independent of price changes. The seller of the commodity is assumed to optimize the trade-off between the received payment for the provision of the service and the loss of flexibility by the reduced selling on the market. The chosen setup allows for closed-form solutions, such that the analysis is of theoretical interest. A potential application is hydropower storage optimization against exogenous electricity prices with the option to enter contracts for providing spinning reserve; spinning reserve is needed to stabilize large-scale power systems. A single-period model is considered, and the storage level of the commodity is bounded from below in expectation. These simplifications allow a closed-form solution of bang-bang type, even under our assumption of an infinite probability space.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 8 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Landir Saviniec, Maristela O. Santos, Alysson M. Costa, Lana M.R. dos Santos〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉High school timetabling problems consist in building periodic timetables for class-teacher meetings considering compulsory and non-compulsory requirements. This family of problems has been widely studied since the 1950s, mostly via mixed-integer programming and metaheuristic techniques. However, the efficient search of optimal or near-optimal solutions is still a challenge for many problems of practical size. In this paper, we investigate mixed-integer programming formulations and a parallel metaheuristic based algorithm for solving high school timetabling problems with compactness and balancing requirements. We propose two pattern-based formulations and a solution algorithm that simultaneously exploits column generation and a team of metaheuristics to build and improve solutions. Extensive computational experiments conducted with real-world instances demonstrate that our formulations are competitive with the best existing high school timetabling formulations, while our parallel algorithm presents superior performance to alternative methods available in the literature.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 5 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Kai A. Konrad〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper studies the attack-and-defence game between a web user and a whole set of players over this user’s ‘valuable secrets.’ The number and type of these valuable secrets are the user’s private information. Attempts to tap information as well as privacy protection are costly. The multiplicity of secrets is of strategic value for the holders of these secrets. Users with few secrets keep their secrets private with some probability, even though they do not protect them. Users with many secrets protect their secrets at a cost that is smaller than the value of the secrets protected. The analysis also accounts for multiple redundant information channels with cost asymmetries, relating the analysis to attack-and-defence games with a weakest link.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 2 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): C. Archetti, D. Feillet, A. Mor, M.G. Speranza〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The dynamic traveling salesman problem with stochastic release dates (DTSP-srd) is a problem in which a supplier has to deliver parcels to its customers. These parcels are delivered to its depot while the distribution is taking place. The arrival time of a parcel to the depot is called its release date. In the DTSP-srd, release dates are stochastic and dynamically updated as the distribution takes place. The objective of the problem is the minimization of the total time needed to serve all customers, given by the sum of the traveling time and the waiting time at the depot. The problem is represented as a Markov Decision Process and is solved through a reoptimization approach. Two models are proposed for the problem to be solved at each stage. The first model is stochastic and exploits the entire probabilistic information available for the release dates. The second model is deterministic and uses an estimation of the release dates. An instance generation procedure is proposed to simulate the evolution of the information to perform computational tests. The results show that a more frequent reoptimization provides better results across all tested instances and that the stochastic model performs better than the deterministic model. The main drawback of the stochastic model lies in the computational time required to evaluate a solution, which makes an iteration of the heuristic substantially more time-consuming than in the case where the deterministic model is used.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 28 September 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): María Isabel Hartillo-Hermoso, Haydee Jiménez-Tafur, José María Ucha-Enríquez〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A new exact algorithm for bi-objective linear integer problems is presented, based on the classic ϵ-constraint method and algebraic test sets for single-objective linear integer problems. Our method provides the complete Pareto frontier 〈math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si19.svg"〉〈mi mathvariant="bold-script"〉N〈/mi〉〈/math〉 of non-dominated points and, for this purpose, it considers exactly 〈math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si5.svg"〉〈mrow〉〈mo〉|〈/mo〉〈mi mathvariant="bold-script"〉N〈/mi〉〈mo〉|〈/mo〉〈/mrow〉〈/math〉 single-objective problems by using reduction with test sets instead of solving with an optimizer. Although we use Gröbner bases for the computation of test sets, which may provoke a bottleneck in principle, the computational results are shown to be promising, especially for unbounded knapsack problems, for which any usual branch-and-cut strategy could be much more expensive. Nevertheless, this algorithm can be considered as a potentially faster alternative to IP-based methods when test sets are available.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 16 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Qichen Deng, Bruno F. Santos, Richard Curran〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper presents a practical dynamic programming based methodology to optimize the long-term maintenance check schedule for a fleet of heterogeneous aircraft. It is the first time that the long-term aircraft maintenance check schedule is optimized, integrating different check types in a single schedule solution. The proposed methodology aims at minimizing the wasted interval between checks. By achieving this goal, one is also reducing the number of checks over time, increasing aircraft availability and, therefore, reducing maintenance costs, while respecting safety regulations. The model formulation takes aircraft type, status, maintenance capacity, and other operational constraints into consideration. We also validate and demonstrate the proposed methodology using fleet maintenance data from a European airline. The outcomes show that, when compared with the current practice, the number of maintenance checks can be reduced by around 7% over a period of 4 years, while computation time is less than 15 minutes. This could result in saving worth $1.1M–$3.4M in maintenance costs for a fleet of about 40 aircraft and generating more than $9.8M of revenue due to higher aircraft availability.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 14 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Urtats Etxegarai, Eva Portillo, Jon Irazusta, Lucien Koefoed, Nikola Kasabov〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this work, a heuristic as operational tool to estimate the lactate threshold and to facilitate its integration into the training process of recreational runners is proposed. To do so, we formalize the principles for the lactate threshold estimation from empirical data and an iterative methodology that enables experience based learning. This strategy arises as a robust and adaptive approach to solve data analysis problems. We compare the results of the heuristic with the most commonly used protocol by making a first quantitative error analysis to show its reliability. Additionally, we provide a computational algorithm so that this quantitative analysis can be easily performed in other lactate threshold protocols. With this work, we have shown that a heuristic (%60 of 〈em〉endurance running speed reserve〈/em〉), serves for the same purpose of the most commonly used protocol in recreational runners, but improving its operational limitations of accessibility and consistent use.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 13 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): P. Goos, U. Syafitri, B. Sartono, A.R. Vazquez〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Mixture experiments usually involve various constraints on the proportions of the ingredients of the mixture under study. In this paper, inspired by the fact that the available stock of certain ingredients is often limited, we focus on a new type of constraint, which we refer to as an ingredient availability constraint. This type of constraint substantially complicates the search for optimal designs for mixture experiments. One difficulty, for instance, is that the optimal number of experimental runs is not known a priori. The resulting optimal experimental design problem belongs to the class of nonlinear nonseparable multidimensional knapsack problems. We present a variable neighborhood search algorithm as well as a mixed integer nonlinear programming approach to tackle the problem to identify D- and I-optimal designs for mixture experiments when there is a limited stock of certain ingredients, and we show that the variable neighborhood descent algorithm is highly competitive in terms of solution quality and computing time.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 14 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Truong Van Nguyen, Li Zhou, Alain Yee Loong Chong, Boying Li, Xiaodie Pu〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Remanufacturing has received increasing attention from researchers over the last decade. While many associated operational issues have been extensively studied, research into the prediction customer demand for, and the market development of, remanufactured products is still in its infancy. The majority of the existing research into remanufactured product demand is largely based on conventional statistical models that fail to capture the non-linear behaviour of customer demand and market factors in real-world business environments, in particular e-marketplaces. Therefore, this paper aims to develop a comprehensible data-mining prediction approach, in order to achieve two objectives: (1) to provide a highly accurate and robust demand prediction model of remanufactured products; and (2) to shed light on the non-linear effect of online market factors as predictors of customer demand. Based on the real-world Amazon dataset, the results suggest that predicting remanufactured product demand is a complex, non-linear problem, and that, by using advanced machine-learning techniques, our proposed approach can predict the product demand with high accuracy. In terms of practical implications, the importance of market factors is ranked according to their predictive powers of demand, while their effects on demand are analysed through their partial dependence plots. Several insights for management are revealed by a thorough comparison of the sales impact of these market factors on remanufactured and new products.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 9 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): N.H. Chieu, V. Jeyakumar, G. Li〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper we show that a convexifiability property of nonconvex quadratic programs with nonnegative variables and quadratic constraints guarantees zero duality gap between the quadratic programs and their semi-Lagrangian duals. More importantly, we establish that this convexifiability is hidden in classes of nonnegative homogeneous quadratic programs and discrete quadratic programs, such as mixed integer quadratic programs, revealing zero duality gaps. As an application, we prove that robust counterparts of uncertain mixed integer quadratic programs with objective data uncertainty enjoy zero duality gaps under suitable conditions. Various sufficient conditions for convexifiability are also given.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 9 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): C.J. Jagtenberg, A.J. Mason〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Most literature on the ambulance location problem aims to maximize coverage, i.e., the fraction of people that can be reached within a certain response time threshold. Such a problem often has one optimum, but several near-optimal solutions may exist. These may have a similar overall performance but provide different coverage for different regions. This raises the question: are we making ‘arbitrary’ choices in terms of who gets coverage and who does not? In this paper we propose to share time between several good ambulance configurations in the interest of fairness. We argue that the Bernoulli–Nash social welfare measure should be used to evaluate the fairness of the system. Therefore, we formulate a nonlinear optimization model that determines the fraction of time spent in each configuration to maximize the Bernoulli–Nash social welfare. We solve this model in a case study for an ambulance provider in the Netherlands, using a combination of simulation and optimization. Furthermore, we analyze how the Bernoulli–Nash optimal solution compares to the maximum-coverage solution by formulating and solving a multi-objective optimization model.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 9 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Jian Luo, Xin Yan, Ye Tian〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Unsupervised classification is a highly important task of machine learning methods. Although achieving great success in supervised classification, support vector machine (SVM) is much less utilized to classify unlabeled data points, which also induces many drawbacks including sensitive to nonlinear kernels and random initializations, high computational cost, unsuitable for imbalanced datasets. In this paper, to utilize the advantages of SVM and overcome the drawbacks of SVM-based clustering methods, we propose a completely new two-stage unsupervised classification method with no initialization: a new unsupervised kernel-free quadratic surface SVM (QSSVM) model is proposed to avoid selecting kernels and related kernel parameters, then a golden-section algorithm is designed to generate the appropriate classifier for balanced and imbalanced data. By studying certain properties of proposed model, a convergent decomposition algorithm is developed to implement this non-covex QSSVM model effectively and efficiently (in terms of computational cost). Numerical tests on artificial and public benchmark data indicate that the proposed unsupervised QSSVM method outperforms well-known clustering methods (including SVM-based and other state-of-the-art methods), particularly in terms of classification accuracy. Moreover, we extend and apply the proposed method to credit risk assessment by incorporating the T-test based feature weights. The promising numerical results on benchmark personal credit data and real-world corporate credit data strongly demonstrate the effectiveness, efficiency and interpretability of proposed method, as well as indicate its significant potential in certain real-world applications.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 9 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Célia Paquay, Yves Crama, Thierry Pironet〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The problem considered in this work stems from a non-profit organization in charge of door-to-door passenger transportation for medical appointments. Patients are picked up at home by a driver and are then dropped at their appointment location. They may also be driven back home at the end of their appointment. Some patients have specific requirements, e.g., they may require an accompanying person or a wheelchair. Planning such activities gives rise to a so-called dial-a-ride problem. In the present work, it is assumed that the requests assigned to the drivers have been selected, and the transportation plan has been established for the next day. However, in practice, appointment durations may vary due to unforeseen circumstances, and some transportation requests may be modified, delayed or canceled during the day. The aim of this work is to propose a reactive algorithm which can adapt the initial plan in order to manage the disruptions and to take care of as many patients as possible in real-time. The plan should be modified quickly when a perturbation is observed, without resorting to major changes which may confuse the drivers and the patients. Several recourse procedures are defined for this purpose. They allow the dispatcher to temporarily delete a request, to insert a previously deleted request, or to permanently cancel a request. Simulation techniques are used to test the approach on randomly generated scenarios. Several key performance indicators are introduced in order to measure the impact of the disruptions and the quality of the solutions.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 8 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Noémie Le Carrer, Scott Ferson, Peter L. Green〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper describes a framework that combines decision theory and stochastic optimisation techniques to address tide routing (i.e. optimisation of cargo loading and ship scheduling decisions in tidal ports and shallow seas). Unlike weather routing, tidal routing has been little investigated so far, especially from the perspective of risk analysis. Considering the journey of a bulk carrier between 〈em〉N〈/em〉 ports, a shipping decision model is designed to compute cargo loading and scheduling decisions, given the time series of the sea level point forecasts in these ports. Two procedures based on particle swarm optimisation and Monte Carlo simulations are used to solve the shipping net benefit constrained optimisation problem. The outputs of probabilistic risk minimisation are compared with those of net benefit maximisation, the latter including the possibility of a ‘rule-of-the-thumb’ safety margin. Distributional robustness is discussed as well, with respect to the modelling of sea level residuals. Our technique is assessed on two realistic case studies in British ports. Results show that the decision taking into account the stochastic dimension of sea levels is not only robust in real port and weather conditions, but also closer to optimality than standard practices using a fixed safety margin. Furthermore, it is shown that the proposed technique remains more interesting when sea level variations are artificially increased beyond the extremes of the current residual models.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 8 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Ju Zhao, Yong-Wu Zhou, Zong-Hong Cao, Jie Min〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Consider a two-echelon supply chain consisting of two manufacturers and a dominant retailer, such as big supermarkets like Walmart. Under a consignment contract with revenue sharing, the two manufacturers sell through the retailer two substitutable products whose demands are dependent on their shelf space and sales prices. The two manufacturers may compete horizontally for shelf space and pricing by three scenarios: Nash game, Stackelberg game, and collusion, and play vertically the retailer-Stackelberg game with the retailer. For each of these horizontal scenarios, we present all participators’ equilibrium strategies and their corresponding profits, based on which the impacts of manufacturers’ cost difference and moving sequence are investigated. Additionally, we discuss whether a horizontal collusion among manufacturers occurs when they choose their scenarios and whether centralization is always beneficial for the entire chain under the considered consignment contract. The study reveals the following results: (i) When the manufacturers compete horizontally, the high-cost manufacturer always sets a high-price and less shelf space strategy, while the low-cost manufacturer always adopts a low-price and more shelf space strategy, which is not affected by their moving sequence. If they collude horizontally, it is just reverse. (ii) When the two manufacturers compete horizontally, all participators’ equilibrium strategies and their corresponding profits are significantly influenced by manufacturers’ moving sequence. (iii) A horizontal collusion between the manufacturers can occur only when their cost difference is relatively small; this finding supplements existing literature. (iv) When the cost difference between manufacturers is relatively big, then centralization may be detrimental to the entire chain, which can explain why several supply chains adopt vertical competition strategies in practice. In addition, we find that these results still hold for the limited shelf space scenario and shelf-space limitation enhances the horizontal and vertical competition intensity by increasing shelf space fee.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 7 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): John H. Drake, Ahmed Kheiri, Ender Özcan, Edmund K. Burke〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Hyper-heuristics have emerged as a way to raise the level of generality of search techniques for computational search problems. This is in contrast to many approaches, which represent customised methods for a single problem domain or a narrow class of problem instances. The current state-of-the-art in hyper-heuristic research comprises a set of methods that are broadly concerned with intelligently 〈em〉selecting〈/em〉 or 〈em〉generating〈/em〉 a suitable heuristic in a given situation. Hyper-heuristics can be considered as search methods that operate on lower-level heuristics or heuristic components, and can be categorised into two main classes: heuristic selection and heuristic generation. The term hyper-heuristic was defined in the early 2000s as a 〈em〉heuristic to choose heuristics〈/em〉, but the idea of designing high-level heuristic methodologies can be traced back to the early 1960s. This paper gives a brief history of this emerging area, reviews contemporary hyper-heuristic literature, and discusses recent hyper-heuristic frameworks. In addition, the existing classification of selection hyper-heuristics is extended, in order to reflect the nature of the challenges faced in contemporary research. Unlike the survey on hyper-heuristics published in 2013, this paper focuses only on selection hyper-heuristics and presents critical discussion, current research trends and directions for future research.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 6 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Sana Dahmen, Monia Rekik, François Soumis, Guy Desaulniers〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper, we address a personalized multi-department multi-day shift scheduling problem with a multi-skill heterogeneous workforce where employees can be transferred between departments under some restrictions. The objective is to construct a schedule that minimizes under-coverage, over-coverage, transfer and labor costs. We propose a novel two-stage approach to solve it: the first stage considers an approximate and smaller problem based on data aggregation and produces approximate transfers. The second stage constructs personalized schedules based on the information deduced from the first stage. An exhaustive experimental study is conducted and proves the efficiency of the proposed approach in terms of solution quality and computing times.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 5 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Peter Nystrup, Erik Lindström, Pierre Pinson, Henrik Madsen〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We propose four different estimators that take into account the autocorrelation structure when reconciling forecasts in a temporal hierarchy. Combining forecasts from multiple temporal aggregation levels exploits information differences and mitigates model uncertainty, while reconciliation ensures a unified prediction that supports aligned decisions at different horizons. In previous studies, weights assigned to the forecasts were given by the structure of the hierarchy or the forecast error variances without considering potential autocorrelation in the forecast errors. Our first estimator considers the autocovariance matrix within each aggregation level. Since this can be difficult to estimate, we propose a second estimator that blends autocorrelation and variance information, but only requires estimation of the first-order autocorrelation coefficient at each aggregation level. Our third and fourth estimators facilitate information sharing between aggregation levels using robust estimates of the cross-correlation matrix and its inverse. We compare the proposed estimators in a simulation study and demonstrate their usefulness through an application to short-term electricity load forecasting in four price areas in Sweden. We find that by taking account of auto- and cross-covariances when reconciling forecasts, accuracy can be significantly improved uniformly across all frequencies and areas.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 5 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Bowei Chen, Jingmin Huang, Yufei Huang, Stefanos Kollias, Shigang Yue〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉While page views are often sold instantly through real-time auctions when users visit websites, they can also be sold in advance via guaranteed contracts. In this paper, we present a dynamic programming model to study how an online publisher should optimally allocate and price page views between guaranteed and spot markets. The problem is challenging because the allocation and pricing of guaranteed contracts affect how advertisers split their purchases between the two markets, and the terminal value of the model is endogenously determined by the updated dual force of supply and demand in auctions. We take the advertisers’ purchasing behaviour into consideration, i.e., risk aversion and stochastic demand arrivals, and present a scalable and efficient algorithm for the optimal solution. The model is also empirically validated with a commercial dataset. The experimental results show that selling page views via both channels can increase the publisher’s expected total revenue, and the optimal pricing and allocation strategies are robust to different market and advertiser types.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 6 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): M. Van Den Eeckhout, M. Vanhoucke, B. Maenhout〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The personnel staffing problem calculates the required workforce size and is determined by constructing a baseline personnel roster that assigns personnel members to duties in order to cover certain staffing requirements. In this research, we incorporate the planning of the duty demand in the staff scheduling problem in order to lower the staffing costs. More specifically, the demand originates from a project scheduling problem with discrete time/resource trade-offs, which embodies additional flexibility as activities can be executed in different modes. In order to tackle this integrated problem, we propose a decomposed branch-and-price procedure. A tight lower and upper bound are calculated using a problem formulation that models the project scheduling constraints and the time-related resource scheduling constraints implicitly in the decision variables. Based upon these bounds, the strategic problem is decomposed into multiple tactical subproblems with a fixed workforce size and an optimal solution is searched for each subproblem via branch-and-price. Fixing the workforce size in a subproblem facilitates the definition of resource capacity cuts, which limit the set of eligible project schedules, decreasing the size of the branching tree. In addition, in order to find the optimal integer solution, we propose a specific search strategy based upon the lower bound and dedicated rules to branch upon the workload generated by a project schedule. The computational results show that applying the proposed search space decomposition and the inclusion of resource capacity cuts lead to a well-performing procedure outperforming different other heuristic and exact methodologies.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 28 September 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Javier Duran-Micco, Evert Vermeir, Pieter Vansteenwegen〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Urban transportation contributes significantly to CO〈sub〉2〈/sub〉 emissions. Public transport systems are a good strategy to reduce these, but the emissions generated by public transport vehicles should not be neglected during the design of the service. The Transit Network Design and Frequency Setting Problem (TNDFSP) has usually been addressed considering only the passengers’ and the operator's point of view. However, we show it is worthwhile to consider also the emissions already during this planning phase.〈/p〉 〈p〉This paper proposes a memetic algorithm to address the bi-objective TNDFSP where both the total travel time and the CO〈sub〉2〈/sub〉 emissions are minimized. The analysis considers a heterogeneous fleet, meaning that buses of different sizes and technologies can be assigned under a budget constraint. The results on benchmark instances show that the proposed memetic algorithm performs as well as state-of-the-art algorithms where CO〈sub〉2〈/sub〉 emissions are not considered. In addition, several experiments are carried out to observe the effect of incorporating emissions and heterogeneous fleet into the model. The heterogeneous fleet allows reducing travel times and emissions at the same time, compared to solutions without a heterogeneous fleet. Moreover, the explicit minimization of CO〈sub〉2〈/sub〉 emissions within a bi-objective framework allows illustrating the trade-off between both objectives. Reductions of about 30% in the emissions can be achieved by increasing the travel time only 1%, while the costs for the operator remain the same. This clearly demonstrates the benefits of considering both the CO〈sub〉2〈/sub〉 emissions and a heterogeneous fleet during the design stage of public transport systems.〈/p〉 〈/div〉

Description: 〈p〉Publication date: Available online 14 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Xingmei Li, Yao-Huei Huang, Shu-Cherng Fang, Youzhong Zhang〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Project portfolio selection problem (PPSP) is usually formulated as a mixed integer polynomial program with cross-product terms. The problem is hard to solve due to the non-convex cross-product terms involved. To find an exact optimal solution, currently available methods adopt different linearization techniques to handle the cross-product terms and then utilize a branch-and-bound scheme for computations. This study proposes an alternative efficient representation for PPSP using fewer continuous variables than the current methods to achieve global optimum. Numerical experiments are presented to demonstrate the effectiveness and efficiency of the proposed method. In addition, the proposed method is integrated with a general binary cut scheme for identifying all alternative solutions for decision makers to consider better options.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 14 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Jesus T. Pastor, C.A. Knox Lovell, Juan Aparicio〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A natural multiplicative efficiency measure for the Constant Returns to Scale proportional directional distance function (pDDF) is derived, relating its associated linear program to that of the well-known output-oriented radial efficiency measurement model. Based on this relationship, a traditional CCD (Caves, Christensen and Diewert) Malmquist index is introduced to show that, when it is based on the new efficiency measure associated with the pDDF, rather than on a radial efficiency measure associated with an oriented distance function, it becomes a Total Factor Productivity (TFP) index. This constitutes a new result, because heretofore the traditional CCD Malmquist index has not been considered a TFP index. Additionally, a new decomposition of the CCD Malmquist index is proposed that expresses productivity change as the ratio of two components, productivity change due to output change in the numerator and productivity change due to input change in the denominator. In an Appendix the efficiency measure is extended to include any returns to scale pDDF.〈/p〉〈/div〉

Description: 〈p〉Publication date: 1 January 2020〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research, Volume 280, Issue 1〈/p〉 〈p〉Author(s): 〈/p〉

Description: 〈p〉Publication date: Available online 12 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Jianming Dong, Ruyan Jin, Taibo Luo, Weitian Tong〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We investigate the approximability of the 〈em〉m〈/em〉 parallel two-stage flow-shop (mP2FS) problem, where a set of jobs is scheduled on the multiple identical two-stage flow-shops to minimize the 〈em〉makespan〈/em〉, i.e., the finishing time of the last job. Each job needs to be processed non-preemptively on one flow-shop without switching to the other flow-shops. This problem is a hybrid of the classic parallel machine scheduling and two-stage flow-shop scheduling problems. Its strong NP-hardness follows from the parallel machine scheduling problem when the number of machines is part of the input. Our main contribution is a polynomial-time approximation scheme (PTAS) for the mP2FS problem when the number of shops is part of the input, which improves the previous best approximation algorithm of a ratio 〈math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si6.svg"〉〈mrow〉〈mo〉(〈/mo〉〈mn〉2〈/mn〉〈mo linebreak="goodbreak"〉+〈/mo〉〈mi〉ϵ〈/mi〉〈mo〉)〈/mo〉〈/mrow〉〈/math〉. Owing to the strong NP-hardness, our PTAS achieves the best possible approximation ratio.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 12 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Luca E. Schäfer, Tobias Dietz, Nicolas Fröhlich, Stefan Ruzika, José R. Figueira〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We investigate the single-source-single-destination “shortest” path problem in directed, acyclic graphs with ordinal weighted arc costs. We define the concepts of ordinal dominance and efficiency for paths and their associated ordinal levels, respectively. Further, we show that the number of ordinally non-dominated path vectors from the source node to every other node in the graph is polynomially bounded and we propose a polynomial time labeling algorithm for solving the problem of finding the set of ordinally non-dominated path vectors from source to sink.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 9 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Sally Giuseppe Arcidiacono, Salvatore Corrente, Salvatore Greco〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The level dependent Choquet integral has been proposed to handle decision making problems in which the importance and the interaction of criteria may depend on the level of the alternatives’ evaluations. This integral is based on a level dependent capacity, which is a family of single capacities associated to each level of evaluation for the considered criteria. We present two possible formulations of the level dependent capacity where importance and interaction of criteria are constant inside each one of the subintervals in which the interval of evaluations for considered criteria is split or vary with continuity inside the whole interval of evaluations. Since, in general, there is not only one but many level dependent capacities compatible with the preference information provided by the Decision Maker, we propose to take into account all of them by using the Robust Ordinal Regression (ROR) and the Stochastic Multicriteria Acceptability Analysis (SMAA). On one hand, ROR defines a necessary preference relation (if an alternative 〈em〉a〈/em〉 is at least as good as an alternative 〈em〉b〈/em〉 for all compatible level dependent capacities), and a possible preference relation (if 〈em〉a〈/em〉 is at least as good as 〈em〉b〈/em〉 for at least one compatible level dependent capacity). On the other hand, considering a random sampling of compatible level dependent capacities, SMAA gives the probability that each alternative reaches a certain ranking position as well as the probability that an alternative is preferred to another. A real-world decision problem on rankings of universities is provided to illustrate the proposed methodology.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 11 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Eberhard Feess, Christian Grund, Markus Walzl, Ansgar Wohlschlegel〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We investigate the choice between posted prices and auctions of competing sellers with private valuations. Assuming that buyers face higher hassle costs in auctions, we show the existence of monotone pure strategy equilibria where sellers offer posted prices rather than auctions if and only if they have a sufficiently high reservation value. Posted prices sell with lower probability but yield a larger revenue in case of trade. Using an empirical strategy to compare revenues of posted prices and auctions that takes selling probabilities explicitly into account, we find our theoretical predictions supported by data from eBay auctions on ticket sales for the EURO 2008 European Football Championship.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 8 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Sven Boge, Sigrid Knust〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper, we consider the parallel stack loading problem (PSLP) with the objective to minimize the number of reshuffles in the retrieval stage. Since in the PSLP the incoming items have to be stored according to a fixed arrival sequence, some reshuffles cannot be avoided later on. We study two surrogate objective functions (number of unordered stackings, number of badly placed items) to estimate the number of reshuffles and compare them theoretically as well as in a computational study. For this purpose, MIP formulations and a simulated annealing algorithm are proposed.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 6 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Milena Bieniek〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Inventories are ubiquitous in nature and inventory control is a crucial activity undertaken in the supply chain (SC) by a company’s management. The Vendor Managed Inventory (VMI) contract has become a common technique for supply chain management (SCM) since the 1980’s. In this technique, the decision about how much inventory to hold is made by the vendor. In the paper, we consider VMI with consignment (VMCI). Consignment is a frequently used form of business arrangement, in which the vendor retains the ownership of the inventory and gets paid by the retailer on actual units sold. Under VMCI, decisions are made in two steps. In the first step, the vendor specifies a consignment price and an order quantity with the objective to maximize the vendor’s expected profit. In the second step, the retailer chooses a retail price which maximizes the retailer’s expected profit. The customer demand is assumed to be stochastic, additive and price–sensitive. Additive uncertainty can produce negative demand realizations, which may occur in adverse market conditions. We prove that in this case an optimal and possibly non–unique solution to VMCI exists. We calculate closed–form formulas for optimal quantities for uniformly distributed demand. Finally, we demonstrate our approach through a numerical example and we show that the imposition of a non–negativity constraint can cause a higher vendor’s expected profit.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 7 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): M.N. López-García, J.E. Trinidad-Segovia, M.A. Sánchez-Granero, I. Pouchkarev〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper, a new long-term memory factor for extending the well-known Fama and French model is proposed and discussed thoroughly. The new long-term memory factor is based on the Hurst exponent and is calculated using the fractal dimension (FD) algorithm. The relevance of the new factor is illustrated using a sample of 1500 largest U.S. companies from different sectors.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 6 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Jinjiang Yuan, C.T. Ng, T.C.E. Cheng〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper, we study multi-agent scheduling with release dates and preemption on a single machine, where the scheduling objective function of each agent to be minimized is regular and of the maximum form (max-form). The multi-agent aspect has three versions, namely ND-agent (multiple agents with non-disjoint job sets), ID-agent (multiple agents with an identical job set), and CO-agent (multiple competing agents with mutually disjoint job sets). We consider three types of problems: The first type (type-1) is the constrained scheduling problem, in which one objective function is to be minimized, subject to the restriction that the values of the other objective functions are upper bounded. The second type (type-2) is the weighted-sum scheduling problem, in which a positive combination of the objective functions is to be minimized. The third type (type-3) is the Pareto scheduling problem, for which we aim to find all the Pareto-optimal points and their corresponding Pareto-optimal schedules. We show that the type-1 problems are polynomially solvable, and the type-2 and type-3 problems are strongly 〈em〉NP〈/em〉-hard even when all jobs’ release dates are zero and processing times are one. When the number of the scheduling criteria is fixed and they are all lateness-like, such as minimizing 〈em〉C〈/em〉〈sub〉max〈/sub〉, 〈em〉F〈/em〉〈sub〉max〈/sub〉, 〈em〉L〈/em〉〈sub〉max〈/sub〉, 〈em〉T〈/em〉〈sub〉max〈/sub〉, and 〈em〉WC〈/em〉〈sub〉max〈/sub〉, where 〈em〉WC〈/em〉〈sub〉max〈/sub〉 is the maximum weighted completion time of the jobs, the type-2 and type-3 problems are polynomially solvable. To address the type-3 problems, we develop a new solution technique that guesses the Pareto-optimal points through some elaborately constructed schedule-configurations.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 6 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Zhuo Jin, Guo Liu, Hailiang Yang〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In this paper, we consider the optimal consumption and investment strategies for households throughout their lifetime. Risks such as the illiquidity of assets, abrupt changes of market states, and lifetime uncertainty are considered. Taking the effects of heritage into account, investors are willing to limit their current consumption in exchange for greater wealth at their death, because they can take advantage of the higher expected returns of illiquid assets. Further, we model the liquidity risks in an illiquid market state by introducing frozen periods with uncertain lengths, during which investors cannot continuously rebalance their portfolios between different types of assets. In liquid market, investors can continuously remix their investment portfolios. In addition, a Markov regime-switching process is introduced to describe the changes in the market’s states. Jumps, classified as either moderate or severe, are jointly investigated with liquidity risks. Explicit forms of the optimal consumption and investment strategies are developed using the dynamic programming principle. Markov chain approximation methods are adopted to obtain the value function. Numerical examples demonstrate that the liquidity of assets and market states have significant effects on optimal consumption and investment strategies in various scenarios.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 2 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Bruno S. Vieira, Sérgio F. Mayerle, Lucila M.S. Campos, Leandro C. Coelho〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The operation of Water Distribution Systems (WDS) is often complex, especially when considering the changes in tariffs throughout the day. The cost of energy in these systems can reach 30% of total operating costs and its careful management can represent increased efficiency. The optimization of WDS scheduling operation appears as an effective method to reduce operating costs while ensuring a good service level to the population. In this paper we propose a new linear relaxation for a non-linear integer programming formulation for WDS in order to optimize its operation costs. This study makes five main contributions. First, our formulation includes new aspects related to the state of the system when the tanks are full, that were not considered before in mathematical programming models. Second, our linearization technique includes a variable number of breakpoints, resulting in significantly fewer binary variables for a given error level. Third, our relaxation reduces the search space of the solutions. Fourth, we have outperformed the best results for three benchmark instances from the literature. Lastly, we also provide a larger new real-life instance with specific conditions of energy tariffs, obtained from the WDS from the city of Florianópolis, southern Brazil, significantly outperforming the current solution employed by the utility provider.〈/p〉〈/div〉

Description: 〈p〉Publication date: Available online 2 August 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research〈/p〉 〈p〉Author(s): Antoine Mandel, Xavier Venel〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We propose an analytical approach to the problem of influence maximization in a social network where two players compete by means of dynamic targeting strategies. We formulate the problem as a two-player zero-sum stochastic game. We prove the existence of the uniform value: if the players are sufficiently patient, both can guarantee the same mean-average opinion without knowing the exact length of the game. Furthermore, we put forward some elements for the characterization of equilibrium strategies. In general, players must implement a trade-off between a forward-looking perspective, according to which they aim to maximize the future spread of their opinion in the network, and a backward-looking perspective, according to which they aim to counteract their opponent’s previous actions. When the influence potential of players is small, we describe an equilibrium through a one-shot game based on eigenvector centrality.〈/p〉〈/div〉

Description: 〈p〉Publication date: 1 May 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 European Journal of Operational Research, Volume 274, Issue 3〈/p〉 〈p〉Author(s): Mauro Dell’Amico, Maxence Delorme, Manuel Iori, Silvano Martello〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We consider the multiple knapsack problem, that calls for the optimal assignment of a set of items, each having a profit and a weight, to a set of knapsacks, each having a maximum capacity. The problem has relevant managerial implications and is known to be very difficult to solve in practice for instances of realistic size. We review the main results from the literature, including a classical mathematical model and a number of improvement techniques. We then present two new pseudo-polynomial formulations, together with specifically tailored decomposition algorithms to tackle the practical difficulty of the problem. Extensive computational experiments show the effectiveness of the proposed approaches.〈/p〉〈/div〉