Stochastic Programming Model Research Articles

Optimal design and robust operational management of regional bioethanol supply chain with various technological choices and uncertainty fusions

Bioethanol has emerged as a promising alternative to fossil fuels, but its commercialization is hindered by high costs and uncertainties surrounding feedstock supply and policies. To address these challenges, a two-stage stochastic robust programming model is developed for regional biorefineries planning and supply chain management with various technological choices and uncertainty fusions. The model maximizes total profits by optimizing site selection, alternative technology combinations, feedstock inventories, and accounting for uncertain and seasonally varying biomass supply. This method helps decision-makers achieve risk-aversion robust optimal solutions, tailored for a case study in Guangdong, China. The results suggest that installing 8 biorefineries using the LHW-SSF-A technique combination would be optimal. The desired feedstock warehouse capacity would be 358.48 ktons, with low utilization closely tied to seasonal crop harvests. Risk-averse decision makers would take conservative production strategies, potentially reducing investment in biorefineries, while market prices and subsidies could encourage fewer conservative investors.

Integrated optimization of facility location, casualty allocation and medical staff planning for post-disaster emergency response

PurposeDisaster management and humanitarian logistics (HT) play crucial roles in large-scale events such as earthquakes, floods, hurricanes and tsunamis. Well-organized disaster response is crucial for effectively managing medical centres, staff allocation and casualty distribution during emergencies. To address this issue, this study aims to introduce a multi-objective stochastic programming model to enhance disaster preparedness and response, focusing on the critical first 72 h after earthquakes. The purpose is to optimize the allocation of resources, temporary medical centres and medical staff to save lives effectively.Design/methodology/approachThis study uses stochastic programming-based dynamic modelling and a discrete-time Markov Chain to address uncertainty. The model considers potential road and hospital damage and distance limits and introduces an a-reliability level for untreated casualties. It divides the initial 72 h into four periods to capture earthquake dynamics.FindingsUsing a real case study in Istanbul’s Kartal district, the model’s effectiveness is demonstrated for earthquake scenarios. Key insights include optimal medical centre locations, required capacities, necessary medical staff and casualty allocation strategies, all vital for efficient disaster response within the critical first 72 h.Originality/valueThis study innovates by integrating stochastic programming and dynamic modelling to tackle post-disaster medical response. The use of a Markov Chain for uncertain health conditions and focus on the immediate aftermath of earthquakes offer practical value. By optimizing resource allocation amid uncertainties, the study contributes significantly to disaster management and HT research.

A two-stage stochastic programming approach for dynamic OD estimation using LBSN data

Estimating origin–destination (OD) demand is essential for urban transport management and traffic control systems. With the ubiquity of smartphones, location based social networks (LBSN) data has emerged as a new rich data source with broad urban spatial and temporal coverage highly suitable for OD estimation. Its nature of confirmed trip purpose (activity) and activity chain host makes it more advantageous than other data (e.g., household travel surveys and traffic network detection). On the other hand, LBSN data is a more direct and accurate representation of demand patterns and can remove the significant burden of developing traffic models and estimating simulation-based objective functions. However, thus far, most LBSN-based estimation models only focus on static (day-level) OD estimation, making less use of those characteristics. To this end, this paper establishes a two-stage stochastic programming (TSSP) framework integrating the activity chains to model activity-level dynamic mobility flows using LBSN data. The first-stage model aims to minimize the errors introduced by the inter-zone OD flows alongside the expected errors of the check-in patterns. The second-stage model attempts to minimize the errors produced by the considered check-in pattern scenarios. Markov chain Monte Carlo (MCMC) sampling is used to generate plausible check-in scenarios. A generalized Benders decomposition (GBD) algorithm is presented to solve the two-stage stochastic programming model. We conduct the experiments on the case study of Tokyo, Japan, under the employment of the generalized least squares (GLS) estimator. The results show that algorithm convergence can be guaranteed within several iterations. The approach can provide satisfactory estimations of check-in patterns, zonal production and attraction, and OD flows. Furthermore, multiple objective function states are tested for evaluating the completeness of the proposed framework and exploring its potential for simplification and extension. Incorporating specific penalty terms into the objective function also provides a way to verify the reliability of the two-stage structure and validate the effectiveness of the model. Finally, we discuss the model enhancement from the perspectives of online OD estimation by integrating with LBSN simulations, network-wide OD extrapolation using appropriate scaling methods, and removing the user-side data requirement by leveraging activity chain modeling. The proposed framework provides a novel and effective approach to OD demand estimation using LBSN data.

Optimization of crude oil operations scheduling by applying a two-stage stochastic programming approach with risk management

This paper focuses on the problem of crude oil operations scheduling carried out in a system composed of a refinery and a marine terminal, considering uncertainty in the arrival date of the ships that supply the crudes. To tackle this problem, we develop a two-stage stochastic mixed-integer nonlinear programming (MINLP) model based on continuous-time representation. Furthermore, we extend the proposed model to include risk management by considering the Conditional Value-at-Risk (CVaR) measure as the objective function, and we analyze the solutions obtained for different risk levels. Finally, to evaluate the solution obtained, we calculate the Expected Value of Perfect Information (EVPI) and the Value of the Stochastic Solution (VSS) to assess whether two-stage stochastic programming model offers any advantage over simpler deterministic approaches.

Privacy-preserving adaptive traffic signal control in a connected vehicle environment

Although Connected Vehicles (CVs) have demonstrated tremendous potential to enhance traffic operations, they can impose privacy risks on individual travelers, e.g., leaking sensitive information about their frequently visited places, routing behavior, etc. Despite the large body of literature that devises various algorithms to exploit CV information, research on privacy-preserving traffic control is still in its infancy. In this paper, we aim to fill this research gap and propose a privacy-preserving adaptive traffic signal control method using partially connected vehicle data. Specifically, we proposed a privacy-preserving mechanism to protect CV data against three types of attacks: CV collusion attacks, database attacks, and inference attacks. The mechanism leverages secure Multi-Party Computation and differential privacy to aggregate individual-level CV data to calculate key traffic parameters without compromising the privacy of CV users. For seamless integration with the privacy-preserving mechanism, we develop a traffic signal optimization model and an arrival rate estimator relying only on aggregated CV data, being applied to both undersaturated and oversaturated traffic conditions. The optimization model is further extended to a stochastic programming problem to explicitly handle the noises added by the privacy-preserving mechanism. Evaluation results show that the linear optimization model preserves privacy with a marginal impact on control performance, and the stochastic programming model can significantly reduce residual queues compared to the linear programming model, with almost no increase in vehicle delay. Overall, our methods demonstrate the feasibility of incorporating privacy-preserving mechanisms in CV-based traffic modeling and control, which guarantees both utility and privacy.

IMPROVEMENT OF THE PROCESS OF CARGO TRANSPORTATION BY ROUTES IN THE DIRECTION TO THE WESTERN BORDERS OF UKRAINE

The paper examines the process of movement of carloads with export goods in the modern conditions of railway transport of Ukraine. In the conditions of a full-scale war, after blocking the ports, the export transportation of goods takes place largely through the western border crossings. The volumes of cargo transportation passing through the western border stations, which have been increasing recently, were studied. Bulk cargo, such as grain, iron ore, ferrous metals, and others, is a significant part of cargo transported in the direction of the western borders to the EU countries or in transit through their territory. The transportation of such cargoes is usually organized as part of route trains, since the large volume of the load allows full trains to be accumulated. The technology of planning the transportation of goods by routes is proposed, which makes it possible to rationally manage the transportation process, provided that trains are dispatched according to a fixed schedule that is consistent with the schedule of the foreign carrier. A model of stochastic programming with optimization of operational costs arising in the process of formation and routing of a route train in the direction of the western borders has been developed. The developed model takes into account the probabilistic nature of the formation time of the route train at the loading station, which is subject to the normal distribution law. To ensure the practical implementation of the model, the technological and regulatory conditions reflected in the system of restrictions are taken into account. The use of the proposed technology will make it possible to avoid the processing of wagons in the process of transportation, increase the productivity of rolling stock by reducing the turnover of wagons, optimize the technology of planning the transportation process, and also increase the predictability of cargo shipments for shippers. In addition, the technology takes into account multiple cases of route delays on a fixed schedule. Determining the amount of train delay is solved using the theory of fuzzy sets. Taking into account the received data on the places of delays and their causes, it is necessary to lay time reserves on the route of freight trains moving according to a fixed schedule.

Carsharing equitable relocation problem: A two-stage stochastic programming approach with learning-embedded endogenous uncertainty in demand

Uncertain demand may exacerbate the imbalance of the supply–demand for a one-way carsharing system and complicate vehicle relocation decisions. To consider the effect of the uncertainty, this study proposes a two-stage stochastic nonlinear programming model, integrating long-term and short-term decisions and maximizing the profit of the carsharing companies. Specifically, in the first stage, tactical decisions of fleet sizing and initial vehicle distribution are determined before the realization of the uncertain demand. Operational decisions of both operator-based and user-based relocation are optimized in the second stage. Moreover, this paper first studies the user-based relocation incentives, which affect the distribution of uncertain demand with an endogenous relationship. A learning-embedded optimization method is introduced to learn such a distribution, enabling the decision-making optimization model to achieve higher performance under the guidance of the demand uncertainty. Second, we envision an equitable relocation issue that considers an uneven distribution of the unsatisfied demand with two different equity criteria measured from the aspects of stations and OD pairs, respectively. Third, the large problem scale, the nonlinear objective function and constraints, and the endogenous demand uncertainty constitute the nontrivial challenges to the feasible solution. For solving the problem efficiently, we linearize the nonlinear terms and develop a dedicated two-phase solution algorithm with a learning-embedded trust-region method in phase I to solve the continuous relaxation problem and a mixed-integer linear programming guided iterative rounding in phase II to obtain the integer solutions of carsharing operations. The solution algorithm adaptively bridges the learning and optimization process via the trust-region method with flexible sample generation. Finally, we conduct numerical experiments based on a real-world one-way carsharing system in Beijing to demonstrate the effectiveness and applicability of the proposed method and reveal some insights for the carsharing service.

Designing a sustainable, resilient, and responsive wheat supply chain under mixed uncertainty: A multi-objective approach

The stability and long-term viability of agri-food supply chains are at risk due to factors such as rapid population growth and global disruptions like the COVID-19 pandemic. This study develops a multi-objective robust fuzzy stochastic programming model to integrate sustainability, resilience, and responsiveness in designing agri-food supply chains operating under uncertain conditions. The sustainability-related objectives of the proposed model aim to reduce the overall cost and negative environmental and social effects. The resilience-related objectives in this model focus on decreasing the node criticality, complexity, and customer dissatisfaction. Additionally, minimizing the total delivery time serves as a measure of responsiveness. In this study, the Torabi and Hassini (TH) interactive fuzzy programming method and the Best-Worst Method (BWM) are applied as a solution approach to handle the multi-objective model. The proposed multi-objective model is applied to a Canadian wheat supply chain, considering its distinct features like blending, wheat quality, weather, and intermodal transport. Moreover, comprehensive sensitivity analyses are carried out to examine the effects of changing the model's parameters on the objective functions. Finally, the performance of the presented model is evaluated by conducting a comparative analysis. The results demonstrate the model's ability to handle hybrid uncertainty and provide tradeoffs between various objectives, which can benefit decision-makers.

Optimization of Communication Base Station Battery Configuration Considering Demand Transfer and Sleep Mechanism under Uncertain Interruption Duration

In the communication power supply field, base station interruptions may occur due to sudden natural disasters or unstable power supplies. This work studies the optimization of battery resource configurations to cope with the duration uncertainty of base station interruption. We mainly consider the demand transfer and sleep mechanism of the base station and establish a two-stage stochastic programming model to minimize battery configuration costs and operational costs. To transform the uncertainty expression in the first stage into a deterministic model, we design the K-Means-SAA algorithm to accelerate problem-solving and to compare it with the SAA algorithm. The case study results indicate that the proposed two-stage stochastic programming model can save 17.02% of the total cost compared to the expected value model. The proposed demand transfer and sleep mechanism can reduce the total cost by 41.92% compared to no mechanism. The results of numerical experiments and sensitivity analysis also verify the superiority of the designed algorithm in terms of running efficiency and solving time. Therefore, the model and algorithm proposed in this work provide valuable application guidance for large-scale base station configuration optimization of battery resources to cope with interruptions in practical scenarios.

Joint relocation and pricing in electric car-sharing systems

In this paper we study the integrated planning problem of determining car-sharing prices between zones of the operating area and routing employees (operators) to relocate cars in preparation for future uncertain demand. We present a novel two-stage integer stochastic programming model for this problem together with a heuristic algorithm, based on Adaptive Large Neighborhood Search (ALNS), to obtain solutions to realistically sized instances. We test the ALNS heuristic on a set of instances generated based on data from a real car-sharing organization and show that it outperforms a commercial solver.

Towards sustainable UAV operations: Balancing economic optimization with environmental and social considerations in path planning

Unmanned Aerial Vehicles (UAVs), i.e., drones, are expected to be widely used in various applications, such as parcel delivery and passenger transport, with the benefits of mitigating traffic congestion and reducing carbon emissions. In this paper, we study a UAV path planning problem under uncertain weather conditions, and design a data-driven dynamic decision support system for multiple types of UAVs. To this end, we categorize all relevant costs into three types, namely, economic, environmental, and social costs, and formulate a nonlinear two-stage stochastic programming model to establish optimal paths for UAV missions under weather uncertainty. We then discretize the nonlinear model and propose a tight linear approximation for the discretized problem to allow for a near real-time implementation. To quantify weather uncertainty, we propose a weather scenario generation algorithm to map ensemble-based weather forecast information to airspace blockage maps. With comprehensive computational studies through simulations, we show that our proposed stochastic approach can lower operating costs by an average of around 6%, where the savings increase as weather conditions become more severe and complex. We also find that, for missions operated by small UAVs, it is not sufficient to determine a path solely based on economic cost minimization, but it should rather be through total cost minimization, which involves environmental and social costs. Considering only the economic cost in the optimization may lead to much higher non-economic costs. However, for missions operated by large UAVs, it is sufficient to determine paths through economic cost optimization, as including environmental and social costs in the optimization process does not result in solutions that are much different from those obtained by considering only the economic costs. For both small and large UAVs, a path established solely through environmental or social cost minimization may not be economically sustainable, as doing so would imply very high economic costs.

Toward equitable grid resilience: operationalizing climate adaptation strategies to mitigate flooding impacts

Disadvantaged communities are disproportionately affected by flooding, exacerbated by climate change. This paper presents a novel framework for incorporating environmental justice into climate adaptation planning of power grids against flooding. A new energy equity metric is introduced with the vision that addressing environmental justice warrants prioritizing disadvantaged communities that have lower risk thresholds. The framework is applied to a levee-protected IEEE standard test system in northern California. The grid performance disturbed due to flooded substations is investigated under current and future climate. The mathematical model of the framework is structured as a two-stage stochastic mixed-integer programming model. This model aims to minimize the equity gap in grid resilience (EGGR) between disadvantaged and non-disadvantaged communities while enhancing the system resilience by reducing the risk of power outages due to flooding. The results show that climate change undermines grid resilience, with disproportionally worse impacts on disadvantaged communities. A significant EGGR is observed that worsens under a changing climate. For adaptation, the optimal placement of distributed energy resources is determined by maximizing the grid resilience to flooding while minimizing EGGR. The proposed framework can equip decision-makers with a robust tool for operationalizing equitable climate adaptation strategies for power grids.

Health-aware coordinate long-term and short-term operation for BESS in energy and frequency regulation markets

The penetration of renewable energy in modern power systems is still increasing. Battery energy storage can rapidly respond to a dispatch order and is expected to provide multiple auxiliary services. However, deep charging cycles have negative impacts on battery health. This paper presents a health-aware long-term operation strategy for lithium-ion battery energy storage participating in the energy and frequency regulation markets. The strategy determines the capacity bounds that can be bid in the energy and frequency regulation markets and updates these bounds every three months, aiming to preserve battery health and increase market revenue. A long-term operational modeling framework is proposed to address the multi-timescale nature, integrating frequency control, energy arbitrage, and the evolution of battery degradation in a holistic model. A nonlinear degradation model is developed to approximate the health impact of main stress factors, which captures the nonlinearity of three-stage capacity degradation process caused by the formation of solid electrolyte interphase film and lithium plating. For intraday operation, a two-scale stochastic programming model is proposed, in which the market bidding and automatic generation control response are simulated in the timescales of one hour and two seconds, respectively. This simulation based method is closer to industrial practice, as it accounts for various factors in BESS operation. For the seasonal update of the capacity allocation strategy, a dynamic programming problem is established and solved based on the nonlinear degradation model and the daily revenue obtained from simulation of intraday operation; the capacity allocation strategy is determined from the renowned Principle of Optimality by Bellman. This two timescale modeling framework captures the interaction between short-term operation strategy and long-term degradation process. Numerical simulations validate that the proposed method can slow down battery degradation and increase lifetime revenue.

Efficient expansion planning of modern multi-energy distribution networks with electric vehicle charging stations: A stochastic MILP model

This paper presents a stochastic Mixed-Integer Linear Programming (MILP) model developed to address the collaborative expansion planning of modern multi-energy distribution networks, taking into account the integration of Electric Vehicle Charging Stations (EVCSs) in the presence of uncertainties. The model is comprehensively formulated to encompass Distributed Generation (DG) resources, Electric Vehicles (EVs), and Capacitor Banks (CBs), alongside traditional expansion options like substations and circuit construction/reinforcement. The essence of this Distribution System Expansion Planning (DSEP) problem formulation lies in its stochastic approach, featuring chance constraints to effectively manage and address uncertainties. The central objective of the model is to minimize the total investment and operational expenditures associated with the DSEP problem over a designated planning horizon. The rigorous application of optimization techniques ensures the convergence of the MILP model, thereby delivering an efficient and reliable solution. The model's effectiveness is initially assessed using an 18-bus system, followed by comprehensive testing on a larger 123- bus system to gauge its robustness. Thorough performance evaluations are conducted across various case studies, encompassing both the 18- bus and 123- bus systems, culminating in the validation of the model's resilience, scalability, and replicability. These case studies encompass deterministic and stochastic methodologies across six distinct scenarios. The numerical outcomes conclusively demonstrate that the DSEP solution, when utilizing the stochastic approach without considering EVCSs, is associated with elevated costs. However, with a strategic allocation strategy for EVCSs, noteworthy reductions in investment expenditures related to substations, lines, and overall costs are notably achieved. The incorporation of uncertainty management within the DSEP problem underscores the flexibility and adaptability intrinsic to the proposed stochastic MILP model. These attributes render it ideally suited to effectively orchestrating the expansion of multi-energy distribution networks in the presence of EVCSs, thereby enhancing its utility and relevance in addressing modern energy distribution challenges.

Manipulating hidden-Markov-model inferences by corrupting batch data

Time-series models typically assume untainted and legitimate streams of data. However, a self-interested adversary may have incentive to corrupt this data, thereby altering a decision maker’s inference. Within the broader field of adversarial machine learning, this research provides a novel, probabilistic perspective toward the manipulation of hidden Markov model inferences via corrupted data. In particular, we provision a suite of corruption problems for filtering, smoothing, and decoding inferences leveraging an adversarial risk analysis approach. Multiple stochastic programming models are set forth that incorporate realistic uncertainties and varied attacker objectives. Three general solution methods are developed by alternatively viewing estimation from frequentist and Bayesian perspectives. The efficacy of each method is illustrated via extensive, empirical testing. The developed methods are characterized by their solution quality and computational effort, resulting in a stratification of techniques across varying problem-instance architectures. This research highlights the weaknesses of hidden Markov models under adversarial activity, thereby motivating the need for robustification techniques to ensure their security.

Two-stage stochastic programming model of US Army aviation allocation of utility helicopters to task forces

US Army aviation units often organize into task forces to meet mission requirements. The manner in which they allocate assets affects their long-term capabilities to provide aviation support. We propose a model to allocate utility helicopters across geographically separated task forces to minimize the total time of flight and unsupported air movement air mission requests (AMRs) by priority level. We model the allocation problem with a two-stage stochastic program, with the first-stage problem allocating a fleet’s helicopter teams to task forces. The stochastic demand for each task force is then revealed. The second-stage US Army aviation air movement operations planning problem is modeled as a stochastic mixed integer linear program (MILP). A practical application uses the air movement operations planning heuristic to solve the second-stage problem at scale and generate an optimal stochastic solution task force allocation. This paper provides evidence for the practical use of the proposed two-stage stochastic programming model for US Army aviation asset allocation by military decision-makers. Furthermore, this research provides a novel first formulation of a stochastic programming dial-a-ride problem with multinode refuel and a sound framework for military aviation asset allocation decision-making.

Day-ahead energy market model for the smart distribution network in the presence of multi-microgrids based on two-layer flexible power management

This study examines the implications of day-ahead energy markets (DA) for flexible power management (FPM) in a smart distribution network (SDN) with multiple microgrids (MMGs). MG sources are coordinated with MG operators (MGOs) in the first layer, and MGOs and SDN sources are coordinated with SDN operators in the second layer. An optimization framework with bilevels is used in the scheme. SDN participation in the wholesale and retail DA energy markets is represented by the upper-level model, equivalent to the second-layer FPM. Based on the network's linear operation and flexibility constraints and active load and source operation models, it minimizes the expected SDN energy cost in the markets mentioned. It has the same formulation as the upper-level problem but concerns MG's participation in the retail DA energy market. This corresponds to the first layer of FPM. After that, a single-level formulation is developed using the Karush-Kuhn-Tucker method, and stochastic programming models uncertainty about load, energy price, renewable generation power, and mobile active load energy demand. In this scheme, two-layer power management between microgrids and the distribution network; simultaneous modelling of economic, operation, and flexibility indices; and simultaneous participation of microgrids and the distribution network in the wholesale and retailer energy markets are the novelties. Ultimately, based on quantitative results, the proposed approach is evaluated in terms of its ability to improve the economic, operational, and flexibility state. The proposed scheme reduces the energy loss in the microgrids and the distribution network by around 43%-67% compared to power flow studies. Voltage drop in the proposed scheme is enhanced by about 40%-47% in the suggested scheme compared to power flow analysis. The economic status of the network is improved by roughly 22% in comparison with power flow studies. Also, the presented scheme can provide almost 100% flexibility conditions; however, it is corresponding to increased energy cost.

Combining deep reinforcement learning and multi-stage stochastic programming to address the supply chain inventory management problem

We introduce a novel heuristic designed to address the supply chain inventory management problem in the context of a two-echelon divergent supply chain. The proposed heuristic advances the current state-of-the-art by combining deep reinforcement learning with multi-stage stochastic programming. In particular, deep reinforcement learning is employed to determine the number of batches to produce, while multi-stage stochastic programming is applied to make shipping decisions. To support further research, we release a publicly available software environment that simulates a wide range of two-echelon divergent supply chain settings, allowing the manipulation of various parameter values, including those associated with seasonal demands. We then present a comprehensive set of numerical experiments considering constraints on production and warehouse capacities under fixed and variable logistic costs. The results demonstrate that the proposed heuristic significantly and consistently outperforms pure deep reinforcement learning algorithms in minimizing total costs. Moreover, it overcomes several inherent limitations of multi-stage stochastic programming models, thus underscoring its potential advantages in addressing complex supply chain scenarios.

Development and Analysis of Optimization Algorithm for Demand-Side Management Considering Optimal Generation Scheduling and Power Flow in Grid-Connected AC/DC Microgrid

The world energy sector is experiencing many challenges, such as maintaining a demand–supply balance with continuous increases in demand, reliability issues, and environmental concerns. Distributed energy resources (DERs) that use renewable energy sources (RESs) have become more prevalent due to environmental challenges and the depletion of fossil fuel reserves. An increased penetration of RESs in a microgrid system facilitates the establishment of a local independent system. However, these systems, due to the uncertainties of RESs, still encounter major issues, like increased operating costs or operating constraint violations, optimal power management, etc. To solve these issues, this paper proposes a stochastic programming model to minimize the total operating cost and emissions and improve the operational reliability with the help of a generalized normal distribution optimization (GNDO). A day-ahead demand response is scheduled, aiming to shift loads to enhance RES utilization efficiency. Demand-side management (DSM) with RESs is utilized, and battery energy storage systems in low-voltage and medium-voltage microgrids are shown. Mathematical formulations of each element in the microgrids were performed. Optimal and consumer-friendly solutions were found for all the cases. Environmental concerns based on the amount of harmful emissions were also analyzed. The importance of demand response is demonstrated vividly. The aim is to optimize energy consumption and achieve optimum cost of operation via DSM, considering several security constraints. A comparative analysis of operating costs, emission values, and the voltage deviation was carried out to prove and justify their potential to solve the optimal scheduling and power flow problem in AC/DC microgrids.

A fast-response mathematical programming approach for delivering disaster relief goods: an earthquake case study

ABSTRACT This paper tackles a complex logistics challenge of disaster management, encompassing warehouse location, pre-disaster inventory planning, routing, and post-disaster relief supply delivery. We establish an iterative process for optimizing relief distribution to shelters. Adaptable warehouse inventory reallocation responds to fluctuating demands, guided by a two-phase mathematical programming approach. In the first phase, a two-stage stochastic programming (TSSP) model determines optimal warehouse and shelter locations and inventory levels. In the subsequent phase, we introduce a mixed-integer programming (MIP) model to minimize the overall delivery time by making routing decisions. To streamline the process, we introduce a novel enumeration algorithm that trims down route options by considering unavailable links, effectively transforming the MIP model into an assignment-based model. This innovation results in a noticeable 74% reduction in solution time. Further efficiency is achieved by developing a branch-and-cut algorithm for swift MIP resolution. A real-world case study confirms the practicality of our approach.