Optimal Decision Research Articles

Multiagent DDOS attack detection model: Optimal trained hybrid classifier and entropy-based mitigation process

ABSTRACT This study proposes a novel multi-agent system designed to detect Distributed Denial of Service (DDoS) attacks, addressing the increasing need for robust cybersecurity measures. The hypothesis posits that a structured multi-agent approach can enhance detection accuracy and response efficiency in DDoS attack scenarios. The methodology involves a five-stage detection model: (1) Preprocessing using a modified double sigmoid normalization technique to eliminate duplicate data; (2) Feature Extraction where raw data and improved correlation-based features, mutual information, and statistical features are identified; (3) Dimensionality Reduction conducted by a reducer agent to streamline the feature set; (4) Classification utilizing Deep Belief Networks (DBN), Bi-LSTM, and Deep Maxout models, with their weights optimally tuned using the hybrid optimization algorithm, WUJSO; and (5) Decision Making by the decision agent to ascertain the presence of attacks, followed by mitigation through modified entropy-based techniques. The results demonstrate that the proposed method achieves a detection accuracy of 0.953 at a learning rate of 90%, significantly outperforming other methods, including Bi-GRU (0.857), DEEP-MAXOUT (0.910), Bi-LSTM (0.865), RNN (0.814), NN (0.894), and DBN (0.761). This research underscores the effectiveness of the multi-agent approach in enhancing DDoS attack detection and mitigation.

Pricing decisions and fairness concerns: supply chain power structures

PurposeThis paper divides the dyadic supply chain into three power structures according to the relative channel power of the supply chain members and consequently examines the optimal supply chain pricing decisions when both suppliers and retailers are concerned with fairness issues.Design/methodology/approachThree models are constructed, namely the Stackelberg game model with the supplier as the leader, the Nash game model with the balance of power and another Stackelberg game model with the retailer as the leader. The equilibrium solutions are solved, and their results are analyzed.FindingsThe retail price of a product increases with an increase in the fairness concerns of the leader in a supply chain in which the supplier or retailer is the leader, while the fairness concerns of the member with less channel power have no effect on the retail price. In a power-balanced supply chain, both suppliers and retailers increase their retail prices as their fairness concerns increase. The relative size of the members’ fairness concerns affects member profits and total supply chain profits.Originality/valueThe main contributions are as follows: First, this paper proposes a new approach to studying supply chain pricing strategy, considering fairness concerns and power structure. Secondly, three game models are constructed. The Nash equilibrium solution is introduced to study the fairness of supply chain participants in pricing decisions and overall supply chain profitability. Finally, the supply chain management theory is expanded by this study on pricing decisions and supply chain performance.

Is there an optimal market withdrawal decision? A simple mathematical model and analysis

Is there an optimal market withdrawal decision? A simple mathematical model and analysis

Multi-Agent Reinforcement Learning based Edge Content Caching for Connected Autonomous Vehicles in IoV

Connected Autonomous Vehicle (CAV) Driving, as a data-driven intelligent driving technology within the Internet of Vehicles (IoV), presents significant challenges to the efficiency and security of real-time data management. The combination of Web3.0 and edge content caching holds promise in providing low-latency data access for CAVs’ real-time applications. Web3.0 enables the reliable pre-migration of frequently requested content from content providers to edge nodes. However, identifying optimal edge node peers for joint content caching and replacement remains challenging due to the dynamic nature of traffic flow in IoV. Addressing these challenges, this article introduces GAMA-Cache, an innovative edge content caching methodology leveraging Graph Attention Networks (GAT) and Multi-Agent Reinforcement Learning (MARL). GAMA-Cache conceptualizes the cooperative edge content caching issue as a constrained Markov decision process. It employs a MARL technique predicated on cooperation effectiveness to discern optimal caching decisions, with GAT augmenting information extracted from adjacent nodes. A distinct collaborator selection mechanism is also developed to streamline communication between agents, filtering out those with minimal correlations in the vector input to the policy network. Experimental results demonstrate that, in terms of service latency and delivery failure, the GAMA-Cache outperforms other state-of-the-art MARL solutions for edge content caching in IoV.

Multi-Objective Optimization and Multi-Criteria Decision-Making Approach to Design a Multi-Tubular Packed-Bed Membrane Reactor in Oxidative Dehydrogenation of Ethane

Multi-Objective Optimization and Multi-Criteria Decision-Making Approach to Design a Multi-Tubular Packed-Bed Membrane Reactor in Oxidative Dehydrogenation of Ethane

Model of selecting an optimal management decision for restoration of fixed assets of oil producing enterprises

Subject. The article explores the problem of oil refineries modernization. Objectives. The study aims at analyzing the options for organization of works on restoration of fixed assets of oil producing enterprises and devising an algorithm for cost optimization. Methods. The study employs methods of complex and comparative analysis. Results. We developed an economic and mathematical model enabling to choose the optimal management decision for the restoration of fixed assets of oil producing enterprises, taking into account geographical and technological factors. Conclusions. The findings may be taken into account when devising development programs for strategically important enterprises in conditions of economic uncertainty.

Multi-Objective Distributionally Robust Optimization for Earthquake Shelter Planning Under Demand Uncertainties

Deciding the locations of shelters and how to assign evacuees to these locations is crucial for effective disaster management. However, the inherent uncertainty in evacuation demand makes it challenging to make optimal decisions. Traditional stochastic or robust optimization models tend to be either too aggressive or overly conservative, failing to strike a balance between risk reduction and cost. In response to these challenges, this research proposes a multi-objective distributionally robust optimization (MODRO) model tailored for shelter location and evacuation allocation. First, an ambiguity set (moment-based or distance-based) is constructed to capture the uncertainty in evacuation demand, reflecting the possible range of outcomes based on demand data from a disaster simulation model. Then, the distributionally robust optimization model considers the “worst-case” distribution within this ambiguity set to minimize construction cost, travel distance, and unmet demand/unused capacity, balancing the trade-off between overly conservative and overly optimistic approaches. The model aims to ensure that shelters are optimally located and evacuees are efficiently allocated, even under the most challenging scenarios. Furthermore, Pareto optimal solutions are obtained using the augmented ε-constraint method. Finally, a case study of Ogu, a wooden density built-up area in Tokyo, Japan, compares the DRO model with stochastic and robust optimization models, demonstrating that the cost obtained by the DRO model is higher than a stochastic model while lower than the worst-case robust model, indicating a more balanced approach to managing uncertainty. This research provides a practical and effective framework for improving disaster preparedness and response, contributing to the resilience and safety of urban populations in earthquake-prone areas.

Fuzzy‐Markov Routing Policy and Priority‐Based Load Balancing for Cognitive Radio Ad Hoc Networks

ABSTRACTA cognitive radio ad hoc network (CRAHNs) is an infrastructure‐free network containing a cognitive radio, which can use the spectrum resources effectively through an adaptive change of parameter settings and decision‐making. Hence, it enhances the spectrum efficacy for satisfying the growing mobile traffic requests. In multihop CRAHNs, in order to ensure a reliable transmission and energy efficiency, the routing protocol should consider the metrics of channel quality, link stability, and energy consumption. In this paper, a Fuzzy‐Markov routing policy and priority‐based load balancing algorithm (FMRP‐PLB) for CRAHNs is designed. In this protocol, a combined routing metric is derived using delay, energy rate, and link expiration time metrics. For modeling the channel state, a Fuzzy‐Markov decision model is designed, in which the channel states and spectrum availabilities of SU are modeled as state transition matrices. Then, optimum routing decisions are made using the combined routing metric and the predicted transmission success probability. A load balancing metric (LBM) is derived in terms of available link capacity, remaining buffer size of node, and back‐off delay. During data transmission, based on the traffic priority and measured levels of LBM, the optimum paths are selected. Simulation results using NS2 suggest that FMRP‐PLB attains an increased packet delivery ratio with reduced delay and energy consumption.

Pricing and platform coupon decisions of a hybrid channel supply chain under different power structures

With the rapid development of e-commerce and the increasing popularity of coupons, we investigate the issuing problem of platform coupons in a hybrid channel supply chain with a manufacturer, a physical retailer and an online platform. Based on game models, we derive the optimal pricing and platform coupon decisions under different power structure. We find that without platform coupons the manufacturer and the retailer obtain more profit if they have more power, while the platform and the system attain the same profit regardless of power structure. With platform coupons, three members obtain more profit if they have more power, while the system obtains more profit if the retailer has more power. If the manufacturer obtains more (less) power than the platform, issuing the platform coupons cannot (can) always attract more online purchase. Moreover, issuing the platform coupons cannot result in the “all-win” outcomes for the members under any power structure. Finally, if the manufacturer obtains more (less) power than the platform, issuing the platform coupons cannot (can) always optimize the chain-wide performance. This paper provides advice for the enterprises to properly use the coupon tool to maximize profit under different power structures.

CEO Inside Debt and Corporate Investment Efficiency

ABSTRACTThis paper investigates whether and how the chief executive officer (CEO)'s inside debt holding influences a firm's investment policy. Our evidence on a sample of US S&P 1500 firms indicates that CEO inside debt can significantly improve investment efficiency and help to overcome both over‐investment and under‐investment, suggesting that the CEO's higher positions of inside debt can lead to optimal investment decisions. The results are robust when we use PSM, entropy balancing, the Heckman two‐stage method and the lagged value of the independent variables to mitigate the endogeneity concerns and other robustness checks. Furthermore, we document that the effectiveness of the CEO inside debt on investment is not sensitive to agency problems, as well as financial constraints. Overall, we confirm the importance of investigating the effect of managerial inside debt holdings on corporate capital investment and provide new insights into how inside debt mitigates agency problems.

Optimum pricing decision for the retailer with warranty and price-dependent variable demand

Buying inventory on a credit basis is an effective pricing plan, and trade credit is a popular component of market transactions that increase demand. In a dynamic situation, most businesses offer different rewards and services to their consumers under certain terms and conditions during commodity sales. The accompanying companies provide a warranty period facility to increase consumer demand for products. The holding cost often increases over time. This study determines a production model wherein the production rate is proportionate to the price-warranty period based on the demand rate with time-dependent holding cost and trade credit policy. This work leads to three vital conventions: (I) the rate of product demand is considered to be price-warranty period dependent; (II) the non-linear function is considered for the rate of replacement failure, wherein the capital of the manufacturer depends on the warranty period; and (III) the rate of inflation is constant; moreover, the present time value of money also measured. This study aims to find the selling price, cycle time, and warranty period by using an algorithm to optimize the total profit function of the manufacturer. The results are validated by solving three numerical examples with their graphical representation based on different situations and the major parameters, and sensitivity analysis is analyzed with important decision-making implications.

Design of Energy Management Strategy for Integrated Energy System Including Multi-Component Electric–Thermal–Hydrogen Energy Storage

To address the challenges of multi-energy coupling decision-making caused by the complex interactions and significant conflicts of interest among multiple entities in integrated energy systems, an energy management strategy for integrated energy systems with electricity, heat, and hydrogen multi-energy storage is proposed. First, based on the coupling relationship of electricity, heat, and hydrogen multi-energy flows, the architecture of the integrated energy system is designed, and the mathematical model of the main components of the system is established. Second, evaluation indexes in three dimensions, including energy storage device life, load satisfaction rate, and new energy utilization rate, are designed to fully characterize the economy, stability, and environmental protection of the system during operation. Then, an improved radar chart model considering multi-evaluation index comprehensive optimization is established, and an adaptability function is constructed based on the sector area and perimeter. Combined with the operation requirements of the electric–thermal–hydrogen integrated energy system, constraint conditions are determined. Finally, the effectiveness and adaptability of the strategy are verified by examples. The proposed strategy can obtain the optimal decision instructions under different operation objectives by changing the weight of evaluation indexes, while avoiding the huge decision space and secondary optimization problems caused by multiple non-inferior solutions in conventional optimization, and has multi-scenario adaptability.

Balancing Profitability and Risk: The Role of Risk Appetite in Mitigating Credit Risk Impact

Balancing profitability and risk is a central challenge for financial institutions, especially in the context of managing credit risk. This paper explores the concept of risk appetite as a strategic framework for aligning an institution’s risk-taking activities with its financial objectives. By defining risk appetite in measurable terms, institutions can calibrate their credit risk strategies to ensure optimal decision-making that mitigates potential losses while maintaining profitability. The study examines how risk appetite influences credit decision-making processes, including credit underwriting, portfolio management, and risk-adjusted return assessment. Through an empirical analysis of credit portfolios across varying economic conditions, the paper demonstrates the effectiveness of a well- defined risk appetite in reducing credit losses and enhancing risk-adjusted returns. Moreover, it highlights the importance of integrating risk appetite frameworks with advanced analytics, such as machine learning and stress testing, to dynamically adapt to changing market conditions. The findings provide actionable insights for financial institutions seeking to strike a balance between risk and reward, ultimately contributing to long-term financial stability and growth. Key Words: Risk Appetite, Credit Risk Management, Profitability, Risk-Adjusted Returns, Credit Underwriting, Portfolio Management, Financial Stability, Risk Mitigation Strategies, Stress Testing, Economic Conditions, Risk- Reward Balance, Credit Decision-Making, Financial Institutions

Post-Natural Disasters Emergency Response Scheme Selection: An Integrated Application of Probabilistic T-Spherical Hesitant Fuzzy Set, Penalty-Incentive Dynamic Attribute Weights, and Non-Compensation Approach

This paper presents an innovative methodology for the dynamic emergency response scheme selection (ERSS) problem in post-major natural disasters. It employs a combination of subjective and objective composite weights and the integrated ELECTRE-score approach. The study aims to provide a practical approach for continuously determining optimal decision schemes at various time points during the decision period in the aftermath of significant natural disasters while accommodating evolving real-world scenarios. Firstly, the probabilistic T-spherical hesitant fuzzy set (Pt-SHFS) captures decision-makers’ ambivalence and hesitation regarding diverse evaluation attributes of different schemes. Subsequently, Pt-SHFS is integrated with the best–worst method (BWM) to determine subjective weights, followed by the structured CRITIC method to amalgamate subjective weights and derive the final combination weights of criteria. Additionally, this paper proposes applying a penalty-incentive mechanism to establish dynamic attribute weights during scenario evolution. Furthermore, the ELECTRE-score method, which may fully exploit the advantages of non-compensation situations, is adopted to obtain more reliable dynamic optimal decision outcomes. Consequently, based on these foundations, an integrated dynamic ERSS approach is formulated to determine optimal dynamic emergency response schemes. Finally, a case study on the Gansu Jishishan earthquake, sensitivity analysis, comparative analysis, and continuous analysis are conducted to verify the practicality, stability, and effectiveness of the proposed approach. The result shows that the proposed comprehensive approach can depict variances among experts’ information, dynamically adjust attribute weights in response to evolving scenarios, and assign a score range and a representative score to each scheme at each decision state. Sensitivity and comparative analyses show this model has strong stability and dynamics. Furthermore, the proposed approach can effectively deal with the complex dynamic situation in the earthquake rescue process, such as the secondary collapse of buildings after the earthquake, the damage of materials caused by heavy rain, and the occurrence of aftershocks. The model can continuously optimize decision-making and provide scientific and reliable support for emergency decision-making.

Green Technology Investment and Production Decisions Considering Stochastic Demand and Firms' Decision Preferences Under Cap‐and‐Trade Regulation

ABSTRACTThis paper investigates how finite rational firms optimally respond to market potential under cap‐and‐trade regulation, considering stochastic demand. Using the mean–variance model framework, we develop an optimization model for green technology investment and production by finite rational firms with different decision‐making preferences. The main results indicate: (1) Under stable demand, firms investing in green technology gain more profits and consumer surplus, however, the effectiveness of carbon emission reduction depends on the market potential size. (2) Under stochastic demand, optimal decisions resulting from firms' varying decision preferences systematically deviate from those of risk‐neutral firms and often demonstrate firms' risk‐averse tendencies that lean towards conservatism. As external factors change, there can even be a reversal in technology investment with an increase in demand uncertainty that amplifies this systematic deviation. (3) Taking firms with different decision‐making preferences as the starting point, this study further analyses the distortion effect of the firm's risk attitude on the optimal technology investment and production decisions, enriching the related research on the joint decision‐making risk model of clean production. And it is more pertinent to the complex situation firms face in the real situation with decision‐making. (4) This study provides a theoretical basis and practical reference for the finite rational firms facing green technology investment choices. Considering the risk‐averse characteristics of firms, this study suggests that the government should establish a risk‐sharing mechanism to weaken the deviation of stochastic demand from the optimal decision‐making of firms or implement certain subsidy policies to protect the profit level of firms.

A Polyhedral Study of Multivariate Decision Trees

Decision trees are a widely used tool for interpretable machine learning. Multivariate decision trees employ hyperplanes at the branch nodes to route datapoints throughout the tree and yield more compact models than univariate trees. Recently, mixed-integer programming (MIP) has been applied to formulate the optimal decision tree problem. A key component of most MIP formulations is a specification of how to route datapoints in the tree from the root to the leaves. Our goal is to provide polyhedral characterizations of realizable routings, that is, routings that can be realized using multivariate branching rules. We first focus on shattering inequalities, a class of valid inequalities that can be used to strengthen almost any MIP formulation and that have been shown to be computationally effective. We prove that if all the feature vectors are in general position, then the shattering inequalities defined for the root of the tree are facet defining for the convex hull of the realizable routings. We then show that every facet-defining inequality of a depth one tree involving at least two variables is also facet defining for trees of arbitrary depth. Finally, we show that facet-defining inequalities characterizing realizable routings are also facet-defining for a complete MIP formulation. We validate our theoretical results by performing numerical experiments that specifically exploit shattering inequalities defined at the root node and we observe an improvement in performance for both numerical and categorical data sets, especially for decision trees of intermediate size. Funding: C. Michini gratefully acknowledges funding from the DoD, Air Force [Award FA9550-23-1-0487].

Optimal pricing and collection decisions in a two-period closed-loop supply chain considering channel inconvenience

Improving recovery efficiency is a key concern for collectors in a closed-loop supply chain (CLSC) with remanufacturing, as customers often consider the inconvenience of recycling channels when returning used products. This issue profoundly affects collectors’ capacity to recover materials. In a two-period CLSC with remanufacturing, including a manufacturer and a retailer, we develop game-theoretical models in the centralized and decentralized scenarios and compare the optimal solutions, consumer surplus, social welfare and environmental impact of different models through analytical and numerical analysis. Our aim is to examine firms’ dynamic pricing strategies and collection investment decisions by considering customers’ perception of channel inconvenience. There are four main findings. Firstly, in the centralized model and the retailer collection model, the decision-maker lowers the retail price in the first period. However, in the competitive collection model, the manufacturer and the retailer raise the wholesale and retail prices in the first period, respectively. Secondly, in the retailer collection model, as the recycling revenue increases, the manufacturer, although not involved in collecting, the profit also increases due to the free-rider behavior. Thirdly, in the competitive collection model, when the remanufacturing cost savings is relatively high, the collection investment of the manufacturer is much larger than that of the retailer, resulting in the retailer failing to collect any product and giving up collecting. Finally, the collection competition improves the total collection rate and environmental performance but reduces the profit of the manufacturer and the retailer, as well as the consumer surplus and social welfare. Therefore, we design a two-part tariff contract to coordinate the decentralized model and effectively improve the performance of the supply chain.

Do more food choices lead to bad decisions? A case study in predaceous ladybird beetle, Propylea dissecta.

Understanding why animals choose one food over another is one of the key questions underlying the fields of behaviour ecology. This study aims to test if ladybird beetle, Propylea dissecta Mulsant (Coleoptera: Coccinellidae) can forage selectively for nutrients in order to redress specific nutritional imbalances to maximise their fitness. We hypothesised that the presence of more food choices leads to bad decisions in terms of their food selection which ultimately negatively affects the mating and reproductive parameters of P. dissecta. To test this, we first manipulated the predator's nutritional status by rearing them in five separate dietary groups from first instar larvae to newly emerged adult stage. Thereafter, we tested their food choice between five different foods, i.e. Aphis craccivora Koch, Aphis nerii Boyer de Fonsclombe, conspecific eggs, heterospecific eggs and mixed pollen grains, equidistantly placed in a Petri dish. Based on the food choice of the newly emerged adults, they were reared on the chosen diet for 10 days. Thereafter, adults were paired with their opposite sex (collected from stock culture reared on A. craccivora) and mating and reproductive parameters were recorded. Our results suggested that the variety of food did not affect the food choice of ladybird beetle, P. dissecta. They tend to choose their natural diet, i.e. aphid in each dietary regime. We found that previous dietary regime, i.e. larval dietary regime, significantly influences the mating and reproductive parameters of both the male and female except for the time to commence mating by the male. Food choices of adult beetles were found to significantly influence the time to commence mating, average fecundity and per cent egg viability in males and only mating duration in females. Our findings suggest that P. dissecta consistently made optimal decisions when facing various food choices. They consistently preferred their natural and preferred food choice over others, indicating a strong food selection behaviour.

The energy management strategy of two-by-one combined cycle gas turbine based on dynamic programming

The complexity and nonlinearity of components in large-scale thermal facilities have resulted in a lack of recognized energy management models, and simple rule-based energy management strategies are still the main approach, which reduces their operating efficiency. In this study, a dynamic programming (DP) method for globally optimal power distribution and operation mode decision for two-by-one combined cycle gas turbine is proposed. First, the energy management model of system is established. Then the drum pressure of the heat recovery steam generator, which indicates the thermal energy storage of the system, is chosen as the state variable, while the control variables are the gas turbine power, turbine power and operation mode. In addition, the system response time is considered to re-evaluated the mode switch command. The simulation results show that the DP optimizes the thermal storage management, which allows the gas turbine to run in the high-efficiency operating range for a longer time. The DP-based strategy saves 6.25%, 5.89%, and 4.92% of fuel at initial drum pressure 8MPa, 9MPa, and 10MPa, respectively, compared to the rule-based strategy. The results of this study can be used as a benchmark to evaluate online energy management strategies in future work.

Degradation-infused energy portfolio allocation framework: Risk-averse fair storage participation

This work proposes a novel degradation-infused energy portfolio allocation (DI-EPA) framework for enabling the participation of battery energy storage systems in multi-service electricity markets. The proposed framework attempts to address the challenge of including the rainflow algorithm for cycle counting by directly developing a closed-form of marginal degradation as a function of dispatch decisions. Further, this closed-form degradation profile is embedded into an energy portfolio allocation (EPA) problem designed for making the optimal dispatch decisions for all the batteries together, in a shared economy manner. We term the entity taking these decisions as ‘facilitator’ which works as a link between storage units and market operators. The proposed EPA formulation is quipped with a conditional-value-at-risk (CVaR)-based mechanism to bring risk-averseness against uncertainty in market prices. The proposed DI-EPA problem introduces fairness by dividing the profits into various units using the idea of marginal contribution. Simulation results regarding the accuracy of the closed-form of degradation, effectiveness of CVaR in handling uncertainty within the EPA problem, and fairness in the context of degradation awareness are discussed. Numerical results indicate that the DI-EPA framework improves the net profit of the storage units by considering the effect of degradation in optimal market participation.