International Transactions in Operational Research最新文献

Amidst labor shortages, increasing land prices, and other aspects of the post-pandemic world, our logistics chain must consider new technological alternatives. Automated material handling equipment often requires valuable floor space and specialized layouts. Uncrewed aerial vehicles, popularly called drones, offer a flexible and cost-effective alternative. To explore the potential benefits and challenges of such systems, we used a version of the multi-trip vehicle routing problem with time windows to model two realistic manufacturing environments as use cases. We provide the mixed integer linear programming formulation of this problem and compare results with those of a human-only based system in a detailed simulated environment. Furthermore, as a physical proof-of-concept, we outfitted a commercially available drone with pick-up and carry capabilities. We found that, even with limited drone carrying capacity, there are economic benefits realized from time savings, compared to ground-based material handling systems. Besides these time savings, drones are sustainable as they operate on battery power and do not impair the air quality or the traffic experienced on the ground.

Policymakers tend to presume that small local education agencies (LEAs) are administratively top heavy, but indivisibilities at the classroom level could just as easily lead small LEAs to overuse teachers rather than administrators. This analysis uses an input distance function and administrative data on students, staff, and spending to estimate the technical and allocative efficiency of Texas public school districts. Our results suggest that small districts are no more likely to overuse administrators than to overuse teachers. Once likely determinants of inefficiency are taken into account, there is no relationship between school district size and the degree of allocative inefficiency. As such, our analysis casts doubt on the efficacy of efficiency rules of thumb that are common in public service practice.

The Physical Internet (PI) is a relatively new logistics paradigm defined as a supply chain framework whose physical components are standardized and optimized with the main objective of increasing the supply chain's overall efficiency, resilience, and sustainability. Given the novelty of the PI concept, there is a lack of scientific literature addressing it from a quantitative point of view, although formulating and solving mathematical models representing resilient PI problems are relevant and innovative issues for academics, practitioners, and governments. In this work, we present a multiperiod mixed-integer programming model to design PI-enabled supply chain networks, in which both cost and resilience are optimized. Hyperconnection and multiple actors are considered in the proposed models. A lexicographic method is proposed to solve these models with multiple objectives, which includes a modified version of the hypervolume measure. Both newly designed and adapted benchmark instances are employed to assess our models' performance. We compare this model against a traditional proprietary logistics model and a horizontal collaboration model between two companies. Results show that hyperconnectivity increases resilience by $��5.5�\%�$ and reduces the supply chain network design costs by $��26.4�\%�$. The risk of not satisfying the destination's demands is reduced as well. Furthermore, in the PI context, we propose a minimax model that has been proved to increase cost equity between the considered actors. This model reduces the average difference between the costs of these actors from $��77�\%�$ to $��4�\%�$.

This paper investigates production and procurement strategies in dynamic demand forecasting. It revolves around a two-tiered supply chain that consists of a manufacturer focusing on capacity investment and a retailer responsible for product sales. Within this framework, we propose a novel mixed contract featuring with time flexibility and options. The retailer has the flexibility to place regular and option orders based on demand forecast updates, with uncertain timing for updates. We analyze the capacity response strategy of the manufacturer and the ordering strategy of the retailer, emphasizing the decision on the optimal ordering time point. Our study derives a unique equilibrium solution for optimal production and procurement strategies, crafting the mixed contract parameters to ensure supply chain coordination. Comparisons with pure contract models highlight the consistent value of time flexibility within the mixed contract, while also identifying the pivotal role of options in enhancing supply chain performance within specific contract structures. Numerical experiments examine the sensitivity of the optimal ordering time and evaluate the impact of the contract parameters on the profits of the manufacturer and retailer.

Consider a freight forwarder who obtains cargo space from multiple carriers. Some capacities are reserved long term in advance through allotment contracts, while some are booked on short-term daily basis. A unified mathematical program for both short- and long-term problems under uncertainties is proposed. Demands are modeled using a probability distribution as in the stochastic program, whereas flight travel times belong to an uncertainty set as in robust optimization. In the first stage, the allotments are determined. In the second stage, the shipment demand materializes but the travel time is still unknown, the forwarder determines the daily allocation among flights, given the first-stage allotment. The objective is to minimize the expected total cost in the two stages. We derive a closed-form solution for a special case, propose a heuristic for a general case, and evaluate its performance via numerical experiments, whose historical records came from one of the largest Thai forwarders.

Electric vehicles (EVs) offer significant potential for advancing sustainable environmental goals. However, their widespread adoption has been concentrated in urban areas, raising challenges for interurban travel. In many countries, charging station networks are primarily located within cities, highlighting a key opportunity for expansion to support longer distance journeys. This article addresses the facility location problem for EV charging stations to enable interurban travel. We propose a multiobjective optimization model based on the flow refueling location model with three objectives: maximizing CO₂ emissions reduction, minimizing total costs, and reducing user charging time. The model is solved using an epsilon constraint approach, and Mexico's charging station network is used as a case study. Through computational experiments, various scenarios are evaluated, and a comparative analysis is performed between electric and internal combustion vehicles. Results show that deploying 20 strategically located charging stations could mitigate 3.1 million tons of CO₂, requiring an investment of nearly USD 3.9 million.

Agenda 2030, also known as the Sustainable Development Goals (SDGs), is a set of global goals adopted by all United Nations Member States in 2015 through the 2030 Agenda for Sustainable Development. The SDGs provide a framework to tackle one of the world's most critical challenges: climate change. The 17 interconnected goals address a broad spectrum of social, economic, and environmental issues. The Government of India has embraced “Agenda 2030,” encompassing 17 SDGs alongside various indicators and targets within the sustainable framework. These indicators were instrumental in assessing the current status of the state concerning SDG attainment. Our primary contribution to this research lies in adopting the prescribed methodology to examine the current status of the SDG India Index 3.0. We focused on evaluating major objectives across Indian states, rating them based on their performance. This study proposes a novel approach utilizing the sum-weighted information (SWI) multicriteria decision analysis/making (MCDA/M) framework to assess the progress of Indian states toward achieving the SDGs based on information measures. By integrating quantitative and qualitative indicators, this method offers a comprehensive evaluation of various aspects. The SWI MCDA/M framework enables policymakers and stakeholders to identify key strengths and weaknesses, prioritize areas for improvement, and formulate targeted strategies to accelerate progress toward the SDGs at the state level in India. Through empirical analysis and case studies, this research contributes to enhancing the effectiveness of SDG implementation efforts, fostering development practices, and promoting inclusive growth across Indian states. The findings indicate that Kerala achieved the first rank, with a 4.516 bits information score and Bihar ranks lowest among states in meeting the standards outlined in “Agenda 2030,” with a 5.530 bits information score.

As the complexity of decisions and development technologies increase, decision support systems (DSSs) emerge to support decision-making and improve the system's interaction with decision-makers (DMs). Hence, this study seeks to show the modulations made in the FITradeoff (Flexible and Interactive Trade-off) DSS, highlighting its flexible and interactive nature. These modulations have been made due to findings from behavioral and neuroscience studies conducted on the FITradeoff method and represent an important advance in research in the multicriteria decision-making/aiding field. The study focuses on the main modulation made in the preference modeling process of the FITradeoff method, namely combining two paradigms for modeling preferences: elicitation by decomposition and holistic evaluation. Thus, the study discusses its implementation in the DSS, shows the flexibility this gives DMs, and highlights the two decision strategies presented in the holistic evaluation, that is, selecting the best alternative and eliminating the worst alternative.

Energy consumption and cycle time are two contradictory optimization objectives for the robotic assembly line balancing problem (RALBP). Indeed, minimizing the cycle time leads to choosing the fastest robots, while minimizing the energy consumption leads to choosing the robots with the smallest powers. In the context of RALBP, cycle time minimization has been extensively studied while energy minimization has been much less considered. Studies dealing with simultaneous minimization of the two later are even scarcer. A bi-objective RALBP considering simultaneous minimization of cycle time and energy consumption is studied in this paper. The energy consumption is calculated based on recent papers from the literature. It includes energy consumed during both operation time and idle time. In this paper, a pseudo-polynomial case is solved thanks to an exact algorithm called split. This latter enumerates all Pareto-optimal solutions corresponding to a given giant sequence of operations. Split is then used as a decoder in a metaheuristic operating in a reduced search space where giant sequences encode solutions. An experimental study is performed on instances taken from the literature to test the suggested encoding–decoding scheme. It shows that the suggested approach yields competitive results compared to the literature.

Supply chain resilience has been extensively investigated at the network and firm levels. More granular studies at the level of product supply chain resilience are scarce. In this paper, we examine relationships between product supply chain resilience, firm resilience, and network resilience. We simulate supply chains with two products in different settings of structural and process diversity, connectivity, and flexibility. The methodology is based on discrete-event simulation. The focus of the analysis is on managerial insights. Our main insights show that the resilience of product supply chains depends on the firm and network resilience, and higher firm and network resilience do not always automatically translate into higher resilience at the product level. Managerial implications are discussed and generalized. The outcomes of our study can be used by supply chain and operations managers to improve the resilience of supply chain with consideration of both product and network levels. We contribute to the literature by offering novel insights on the interrelations between firm and network resilience practices and product supply chain resilience.