Search

In this article, we explore the topic of supply chain resilience (SCR) in an open-system context. Systems theory and cybernetics are important methodologies in the analysis of complex networks. However, relatively little is known about their application in the supply chain (SC) operations. Using systems theory and cybernetics for SCR conceptualization, we argue that a systems-theoretical lens can help to extend the state-of-the-art research to describe SCR based on the concept of balancing vulnerability and recoverability over time. Toward this end, we define and test a new resilience measure for SCR using the Bayesian network approach with a compounding function of vulnerability and recoverability. We further propose managerial actions for implementing an open-system view of SCR. To demonstrate the applicability and usefulness of our method, we measure and analyze the resilience levels of a manufacturer. Our results provide useful decision-making support and can be valuable for SC managers seeking to balance risk mitigation and recovery capabilities. The framework and measure developed can also be used to guide firms seeking to analyze the impact of disruptions on SCs through the consideration of both performance degradation and recovery.

While the majority of companies anticipated the negative and severe impacts of the COVID-19 pandemic on the supply chains (SC), most of them lacked guidance on how to model disruptions and their performance impacts under pandemic conditions. Lack of such guidance resulted in delayed reactions, incomplete understanding of pandemic impacts, and late deployment of recovery actions. In this study, we offer a method of modelling and quantifying the SC disruption impacts in the wake of a pandemic. We develop a multi-layer Bayesian network (BN) model that can be used to identify SC disruption triggers and risk events amid the COVID-19 pandemic and quantify the consequences of pandemic disruptions. The unique features of BN, such as forward and backward propagation analysis, are utilised to simulate and measure the impact of different triggers on SC financial performance and business continuity. In this way, we combine resilience and viability SC perspectives and explicitly account for the pandemic settings. The outcomes of this research open a novel theoretical lens on application of BNs to SC disruption modelling in the pandemic setting. Our results can be used as a decision-support tool to predict and better understand the pandemic impacts on SC performance.

Smart microgrids are experiencing an increasing growth due to their economic, social, and environmental benefits. However, the inherent intermittency of renewable energy sources (RESs) and users’ behavior lead to significant uncertainty, which implies important challenges on the system design. Facing this issue, this article proposes a novel robust framework for the day-ahead energy scheduling of a residential microgrid comprising interconnected smart users, each owning individual RESs, noncontrollable loads (NCLs), energy- and comfort-based CLs, and individual plug-in electric vehicles (PEVs). Moreover, users share a number of RESs and an energy storage system (ESS). We assume that the microgrid can buy/sell energy from/to the grid subject to quadratic/linear dynamic pricing functions. The objective of scheduling is minimizing the expected energy cost while satisfying device/comfort/contractual constraints, including feasibility constraints on energy transfer between users and the grid under RES generation and users’ demand uncertainties. To this aim, first, we formulate a min–max robust problem to obtain the optimal CLs scheduling and charging/discharging strategies of the ESS and PEVs. Then, based on the duality theory for multi-objective optimization, we transform the min–max problem into a mixed-integer quadratic programming problem to solve the equivalent robust counterpart of the scheduling problem effectively. We deal with the conservativeness of the proposed approach for different scenarios and quantify the effects of the budget of uncertainty on the cost saving, the peak-to-average ratio, and the constraints’ violation rate. We validate the effectiveness of the method on a simulated case study and we compare the results with a related robust approach. Note to Practitioners —This article is motivated by the emerging need for intelligent demand-side management (DSM) approaches in smart microgrids in the presence of both power generation and demand uncertainties. The proposed robust energy scheduling strategy allows the decision maker (i.e., the energy manager of the microgrid) to make a satisfactory tradeoff between the users’ payment and constraints’ violation rate considering the energy cost saving, the system technical limitations and the users’ comfort by adjusting the values of the budget of uncertainty. The proposed framework is generic and flexible as it can be applied to different structures of microgrids considering various types of uncertainties in energy generation or demand.

This paper proposes a decision support system based on a machine-learned Bayesian network (BN) to predict the success rate of telemarketing calls for long-term bank deposits. Telemarketing is one o...

The problem of frequent itemset mining is one of the more important problems in data mining which has been extensively employed across a wide range of other relevant tasks such as market basket analysis in marketing, or text analysis in text mining applications. The majority of the deterministic frequent itemset mining algorithms which have been proposed in recent years use some sort or another of an optimal data structures to reduce the overall execution time of the algorithm. In this paper, however, we have tried instead to introduce an approximation algorithm which works by converting the problem into a clustering problem where similar transactions are grouped together. Each cluster centroid represents an itemset which may be assumed to be a candidate frequent itemsets. The validity of this assumption is simply verified by calculating the support count of these itemsets. Those who meet the min-support condition are considered to be an actual frequent itemset. As for the remaining itemsets, they are then passed to MAFIA which extract all maximal frequent itemsets therefrom. Experimentations made on several well-known and diverse datasets show that the proposed algorithm performs almost always faster, and in some cases up to 10 times faster, than the existing deterministic algorithms, and all this by retaining up to 95% of its accuracy.

Ports are an integral part of the transportation system and are often susceptible to a diverse range of risks, including natural disasters, malicious cyber-attacks, technological factors, organizational factors, economic factors, and human error. To address the challenges triggered by these diverse risks, this research identifies the basic factors that could enhance the resilience of the port system. After these factors are identified and expressed as different resilience capacities, they are used to quantify the resilience of the port infrastructure by applying a Bayesian network. Quantification of resilience is further analyzed based on different advanced techniques such as forward propagation, backward propagation, sensitivity analysis, and information theory. The formal interpretation of these analyses indicates that maintenance, alternate routing, and manpower restoration are the leading factors contributing to enhancing the resilience of a port infrastructure system under disruptive conditions.

This is the first study that presents a supply chain (SC) resilience measure with the ripple effect considerations including both disruption and recovery stages. SCs have become more prone to disruptions due to their complexity and strategic outsourcing. While development of resilient SC designs is desirable and indeed critical to withstand the disruptions, exploiting the resilience capabilities to achieve the target performance outcomes through effective recovery is becoming increasingly important. More adversely, resilience assessment in multi-stage SCs is particularly challenged by consideration of disruption propagation and its associated impact known as the ripple effect. We theorize a new measure to quantify the resilience of the original equipment manufacturer (OEM) with a multi-stage assessment of suppliers’ proneness to disruptions and the SC exposure to the ripple effect. We examine and test the developed notion of SC resilience as a function of supplier vulnerability and recoverability using a Bayesian network and considering disruption propagation using a real-life case-study in car manufacturing. The findings suggest that our model can be of value for OEMs to identify the resilience level of their most important suppliers based on forming a quadrant plot in terms of supplier importance and resilience. Our approach can be used by managers to identify the disruption profiles in the supply base and associated SC performance degradation due to the ripple effect. Our method explicitly allows to uncover latent, high-risk suppliers to develop recommendations to control the ripple effect. Utilizing the outcomes of this research can support the design of resilient supply networks with a large number of suppliers: critical suppliers with low resilience can be identified and developed.

Resilient supplier selection is a key strategic decision in the context of the supply chain (SC) disruption management. We offer an efficient solution to the resilient supplier selection and optimal order allocation problem. We first show how to compute the likelihood of disruption scenarios for the supplier selection problem using a probabilistic graphical model. That model can capture (i) a large number of disruptive events with no computational burden, and (ii) the dependencies among disruptive events and their impacts on supplier performance, i.e., the ripple effect. We then propose a stochastic bi-objective mixed integer programming model to support the decision-making in how and when to use both proactive and reactive strategies in supplier selection and order allocation. The outcomes of this research, if utilized properly, can benefit suppliers to find the optimal set of operational decisions (e.g., the optimal level of surplus capacity and restorative capacity) that enhance their resilience capabilities. Finally, the proposed model can be utilized as a decision support tool to assist manufacturers in performance assessment of supplier alternatives when costs and resilience are considered simultaneously, which helps to build up both efficient and resilient SC (i.e., to achieve the SC resilience) to ensure the operations continuity. These outcomes can help SC managers organize their disruption risk mitigation efforts with balancing the efficiency and resilience while focusing on critical suppliers and order (re)-allocation that will have a more significant impact on the performance of the SC when disrupted.

This research utilizes Bayesian network to address a range of possible risks to the electrical power system and its interdependent networks (EIN) and offers possible options to mitigate the consequences of a disruption. The interdependent electrical infrastructure system in Washington, D.C. is used as a case study to quantify the resilience using the Bayesian network. Quantification of resilience is further analyzed based on different types of analysis such as forward propagation, backward propagation, sensitivity analysis, and information theory. The general insight drawn from these analyses indicate that reliability, backup power source, and resource restoration are the prime factors contributed towards enhancing the resilience of an interdependent electrical infrastructure system.

Because ports are considered to be the heart of the maritime transportation system, thereby assessing port performance is necessary for a nation’s development and economic success. This study proposes a novel metric, namely, “port performance index (PPI)”, to determine the overall performance and utilization of inland waterway ports based on six criteria, port facility, port availability, port economics, port service, port connectivity, and port environment. Unlike existing literature, which mainly ranks ports based on quantitative factors, this study utilizes a Bayesian Network (BN) model that focuses on both quantitative and qualitative factors to rank a port. The assessment of inland waterway port performance is further analyzed based on different advanced techniques such as sensitivity analysis and belief propagation. Insights drawn from the study show that all the six criteria are necessary to predict PPI. The study also showed that port service has the highest impact while port economics has the lowest impact among the six criteria on PPI for inland waterway ports.

As many large-scale systems age, and due to budgetary and performance efficiency concerns, there is a need to improve the decision-making process for system sustainment, including maintenance, repair, and overhaul (MRO) operations and the acquisition of MRO parts. To help address the link between sustainment policies and acquisition, this work develops a greedy heuristic-based local search algorithm (GHLSA) to provide a system maintenance schedule for multi-component systems, coordinating recommended component maintenance times to reduce system downtime costs, thereby enabling effective acquisition. The proposed iterative algorithm aims to minimize the sum of downtime, earliness and tardiness costs of scheduling, which contains three phases: (1) the construction phase, which uses a heuristic to construct an initial partial solution, (2) an improvement phase, which aims to improve the partial solution generated in the construction phase, and finally, (3) a local search phase, which performs a local search technique to the partial solution found in the improvement phase. The proposed algorithm makes a trade-off between exploration and exploitation of solutions. The experimental results for small (10 jobs) and large size (50 jobs) problems indicate that GHLSA outperforms both genetic algorithm and simulated annealing approaches in terms of solution quality and is similar in terms of efficiency.

Fast charging sites play a crucial role for public acceptance of electric vehicle (EV). Selection of the most sustainable site positively contributes to the life cycle of electric vehicle charging station (EVCS), which requires considering some conflicting criteria. Previous researches mainly focused on utilizing optimization models to deal with EVCS site selection that only accounts for quantitative factors, while this paper proposes a Bayesian Network (BN) model that considers not only quantitative factors but also qualitative (subjective) ones. Based on academic literature and expert judgments, the assessment index for EVCS site selection was mainly made from sustainability point of view, which contains of economic, environmental, and social criteria with a total of eleven sub-criteria. BNs are powerful tools for handling risk assessment and decision making under uncertainty. The developed BN model is validated through sensitivity analysis approach. Finally, different propagation analyses have been performed to make special types of reasoning. This paper provides a new research perspective by considering uncertainty, qualitative and quantitative factors into the site selection assessment, and presents the mainstream penetration of BN as a powerful decision making tool in the context of electrical energy management.

Because ports are considered to be the heart of the maritime transportation system, thereby assessing port performance is necessary for a nation’s development and economic success. This study proposes a novel metric, namely, “port performance index (PPI)”, to determine the overall performance and utilization of inland waterway ports based on six criteria, port facility, port availability, port economics, port service, port connectivity, and port environment. Unlike existing literature, which mainly ranks ports based on quantitative factors, this study utilizes a Bayesian Network (BN) model that focuses on both quantitative and qualitative factors to rank a port. The assessment of inland waterway port performance is further analyzed based on different advanced techniques such as sensitivity analysis and belief propagation. Insights drawn from the study show that all the six criteria are necessary to predict PPI. The study also showed that port service has the highest impact while port economics has the lowest impact among the six criteria on PPI for inland waterway ports.

Relative stopping power (RSP) values of tissues in patients are needed to plan proton beam therapy accurately. Proton CT (pCT) is an alternative imaging method for obtaining more accurate RSP values than by using X-ray CT. This imaging modality gives mostly accurate RSP values but is blurred due to elastic multiple Coulomb scattering. To improve the blurriness of reconstructed pCT images, we have investigated a denoising convolutional neural network trained on known ground RSP values of a digital phantom. In our initial results, with the denoising network receiving pCT images reconstructed with an iterative method as input we observed improved spatial resolution and better RSP accuracy in the output images. The improved images had a higher peak signal-to-noise ratio (PSNR) and significantly improved structural similarity index measure (SSIM). More accurate RSP values with better spatial resolution will pave the way for more widespread adoption of pCT for proton treatment planning.

Total fidelity value index can be used for the assignment of new releves to existing vegetation units and it can be used to refine classifications derived from unsupervised clustering. Diagnostic species is an important concept in vegetation classification. Apart from its usefulness to characterize species niche preferences, the diagnostic species concept is used in vegetation classification: (1) for the assignment of new releves to the vegetation units of an existing classification; (2) to refine vegetation classifications by reassigning releves that sustain the definition of vegetation units. The main aims were to evaluate the relative predictive performance of different statistical fidelity measures for the reassignment of releves to existing vegetation units, and in which cases reassignments improve the quality of the original classification. We took the classifications produced by three commonly used unsupervised classification methods, and all releves were reassigned to the closest vegetation unit according to the total fidelity value index (TFVI), where fidelity value had been calculated using one of eight distinct statistical measures, and according to the frequency-positive fidelity index (FPFI). Classifications obtained after releve reassignments were compared to the initial ones using the Adjusted Rand Index. The quality of all classification solutions, including the initial ones, was evaluated using thirteen different evaluator statistics. The predictive performance of IndVal was the best among all eight fidelity indices in the TFVI framework, and also outperformed FPFI. The TFVI framework based on group-equalized fidelity indices produced better results than other assignment rules in terms of the chosen evaluator statistics. Re-assignments based on IndVal, r, or FPFI produced classifications with the best quality, when combining the results of all evaluators. We conclude that TFVI based on IndVal and r has the best quality for assigning of new releves to existing vegetation units, and it also could be used to refine classifications derived from unsupervised clustering. Consequently, our results reiterate that TFVI, which is new in vegetation sciences, can be a good alternative for FPFI, as the most commonly used in the assignment of vegetation plots (releves), to predefined vegetation types in large datasets.

Filtered back projection (FBP) is the traditional image reconstruction method in x-ray computed tomography (CT). To further improve the quality of FBP-based image reconstruction, often iterative methods are used. But these methods are slow and computationally expensive. In applications where fast and accurate images are needed, FBP may not be sufficient. We propose a very lightweight 3D convolutional neural network (3D CNN) to improve the image quality of the FBP method with minimal computational cost. 3D CNN can be used in time-sensitive situations where iterative methods are not ideal due to time constraints. The method takes advantage of the 3D-nature of the CT data. Popular denoising CNN-based algorithms have been applied to 2D images and were originally designed to correct the effects of 2D artifacts like jpg compression. These approaches fail to utilize the 3D geometry of the CT data and try to improve the 2D CT images individually. The focus of this work was on training and applying a time-saving method to proton CT head data. Proton CT (pCT) requires low-noise reconstruction of relative stopping power (RSP) for proton beam therapy planning and pretreatment verification, where time is of the essence. The proposed fast method successfully improved peak SNR (PSNR) of FBP pCT images and showed a numerical advantage of PSNR for 3D over 2D approaches. In future work, the implementation of existing popular 2D denoising methods in 3D CNN CT reconstruction will be investigated.

This paper deals with the problem of voltage and frequency restoration in droop-controlled inverter-based islanded microgrids (MGs). The problem is solved using a novel tracking consensus secondary control protocol based on the sliding-mode approach, which not only guarantees the exact finite-time restoration among voltages and frequencies of an inverter-based islanded microgrid, but also preserves the active power sharing among distributed generations (DGs). For each distributed generator with no direct access to reference values, a finite-time distributed estimator is locally designed and implemented in each DG unit to provide the reference value of frequency and voltage in a finite time. Lyapunov analysis is employed to verify the associated stability and fast convergence time of the proposed controller. Finally, simulation results are are presented and analyzed to confirm the effectiveness of the proposed approach.

Nowadays, developing coordinated optimal charging strategies for large-scale electric vehicle (EV) fleets is crucial to ensure the reliability and efficiency of power grids. This paper presents a novel fully distributed control strategy for the optimal charging of large-scale EV fleets aiming at the minimization of the aggregated charging cost and battery degradation, while satisfying the EVs' individual load requirements and the overall grid congestion limits. We formulate the optimization problem as a convex quadratic programming problem where all the EVs' decisions are coupled both via the objective function and some grid resource sharing constraints. Based on the distributed waterfilling approach, the proposed resolution algorithm requires a minimal shared information between EVs that communicate only with their neighbors without relying on a central aggregator, thus guaranteeing the EV users' privacy. The performance of the proposed approach is evaluated through numerical experiments to validate its effectiveness in achieving a global optimum while respecting the grid constraints with a favorable computational efficiency.

Data clustering aims to discover the underlying structure of data. it has many applications in data analysis and it is one of the most widely used tools in data mining. DBSCAN is one of the most famous clustering algorithms. its advantages are to identify clusters of various shapes and define the number of clusters. Since DBSCAN is sensitive to its parameters which are e and MinPts, it may perform poorly when the dataset is unbalanced. To solve this problem, this paper proposes a sliding window DBSCAN clustering algorithm that uses Gridding and local parameters for unbalanced data which we will refer to as SW-DBSCAN. The algorithm divides the dataset into several grids. The size and shape of each gird depends on the specimen density specification. Then, for each grid, the parameters are adjusted for local clustering and eventually merging data zones. Experimental results show that this algorithm can help to improve the performance of the DBSCAN algorithm and can deal with arbitrary data and asymmetric data.

We present an approximate algorithm for finding frequent itemsets. The main idea can be described as turning the problem of mining frequent itemsets into a clustering problem. More precisely, we first represent each transaction by a vector using one-hot encoding scheme. Then, by means of mini batch k-means, we group all transactions into a number of clusters. The center of each cluster can be assumed as a potential candidate for a frequent itemset. To test the validity of this assumption, we compute the support of itemsets represented by cluster centers. All clusters that do not meet the minimum support condition will be removed from the set of clusters. As our experiments show, this approximate algorithm can capture more than 90% of all frequent itemsets at a much faster rate than the competing algorithms. Moreover, we show that the execution time of our algorithm is linear.

This paper proposes a novel robust decentralized charging strategy for large-scale EV fleets. The system incorporates multiple EVs as well as inelastic loads connected to the power grid under power flow limits. We aim at minimizing both the overall charging energy payment and the aggregated battery degradation cost of EVs while preserving the robustness of the solution against uncertainties in the price of the electricity purchased from the power grid and the demand of inelastic loads. The proposed approach relies on the so-called uncertainty set-based robust optimization. The resulting charge scheduling problem is formulated as a tractable quadratic programming problem where all the EVs' decisions are coupled via the grid resource-sharing constraints and the robust counterpart supporting constraints. We adopt an extended Jacobi-Proximal Alternating Direction Method of Multipliers algorithm to solve effectively the formulated scheduling problem in a decentralized fashion, thus allowing the method applicability to large scale fleets. Simulations of a realistic case study show that the proposed approach not only reduces the costs of the EV fleet, but also maintains the robustness of the solution against perturbations in different uncertain parameters, which is beneficial for both EVs' users and the power grid.

International audience; The ripple effect could occur when a disruption in supplier base cannot be localised and its consequences propagate downstream the supply chain (SC) adversely affecting the performance. While stress-testing of SC designs and assessment of their vulnerability to disruptions in single-echelon-single-event setting is desirable and indeed critical for some firms, modeling the ripple effect impact in multi-echelon-correlated-events systems is becoming increasingly important. Notably, ripple effect assessment in multi-stage SCs is particularly challenged by consideration of both vulnerability and recoverability capabilities at individual firms in the network. We construct a new model based on integrating Discrete-Time Markov Chain (DTMC) and Dynamic Bayesian Network (DBN) model to quantify the ripple effect. We use DTMC to model the recovery and vulnerability of suppliers. The proposed DTMC model is then equalized with a DBN model in order to simulate the propagation behavior of supplier disruption in the SC. Finally, we propose a metric that quantifies the ripple effect of supplier disruption on manufacturer in terms of total expected utility and service level. The ripple effect measure constructed is examined and tested using two case-studies. The findings suggest that our model can be of value in revealing latent high-risk paths in the SC, analysing the performance impact of both a disruption and its propagation, and prioritizing the contingency and recovery policies.

In the wake of the COVID-19 pandemic, many firms lacked a strategy to cope with disruptions and maintain resiliency. In this study, we develop a measurement method to evaluate the impact of resilience strategies in a multi-stage supply chain (SC) in the presence of a pandemic. For the first time, we propose a method to deduce quantitative resilience assessment from simulation. We implement two resilience strategies, i.e., prepositioning extra-inventory and a backup supplier, and then we simulate its impact on SC resilience and financial performance. The simulation results indicate that the extra inventory leads to a higher resilience than a backup supplier but costs more for the given contextual setting. Finally, we examine the demand fulfillment and observe that the extra-inventory strategy allows for a higher service level, confirming our resilience simulations. We discuss the managerial implications of these findings on the descriptive and predictive analysis levels. Decision-makers can utilize our model and findings to develop a response plan in the occurrence of a pandemic or any long-duration high magnitude disruption. Also, scholars and managers can use our proposed method to measure SC resiliency from simulation in any disruption.

The concept of intermodality has been gaining importance in both freight and passenger transportation across the world. It is essential to identify critical intermodal infrastructures that will help stakeholders prioritize protection initiatives or add necessary redundancy to maximize intermodal network resiliency. In this paper, we identify the critical infrastructures of an integrated intermodal network. We present an optimization model that facilitates all OD-specific traffic flow with optimum cost, utilizing an integrated intermodal network consisting of three surface transportation modes: highway, railway, and waterway. The proposed model considers the congestion effects of each mode and intermodal transfer terminals, which may happen under heavy traffic congestion. We implement a real-world case study to show the critical infrastructures for a given intermodal network of the USA. An integrated way to utilize all surface transportation modes is expected to improve the overall freight flow significantly. Making quick and correct traffic management decisions is a big challenge in a damaged or highly congested transportation network. In addition to criticality analysis, the model will help traffic management decision-making in a disaster by rerouting the traffic in the residual network. The proposed model provides the optimal routing and rerouting paths for the overall freight flow during and after a disaster to enhance surface transportation resilience and security.

This paper presents a real-time demand side management framework based on robust model predictive control (RMPC) for residential smart grids. The system incorporates a number of interconnected smart homes, each equipped with controllable and non-controllable loads, as well as a shared energy storage system (ESS). We aim at minimizing the users’ energy payment and limiting the peak-to-average ratio (PAR) of the energy consumption while taking into account all device/comfort/contractual constraints, specifically the feasibility constraints on energy transferred between users and the power grid in presence of load demand uncertainty. We consider a quadratic cost function for energy bought from the grid. Firstly, the energy price and related constraints of the system are modeled. Then, a min-max robust problem is established to optimally schedule energy under an interval-based uncertainty set. We finally adopt model predictive control (MPC) to solve the resulting robust optimization problem iteratively over a finite-horizon time window based on the receding horizon concept. Moreover, the robustness of the proposed real-time approach against the level of conservativeness of the solution is addressed. The effectiveness of the method is validated through a simulated case study.

In this paper, we propose a novel robust framework for day-ahead energy scheduling of interconnected smart homes with shared energy storage system (ESS), taking into account users’ behavior uncertainty. The objective is minimizing the total energy payment for each user while satisfying the constraint on the feasibility of energy transactions between users and the power grid in presence of data uncertainty. Unlike most existing robust scheduling frameworks that assume a linear cost function for energy purchased from the grid, our design presents a tractable robust optimization scheme to solve the energy scheduling problem with a more realistic quadratic cost function. We model device/comfort constraints as well as contractual obligations imposed by the power grid restricting the users’ energy consumption to a maximum level at each time slot. Thus, in our problem, uncertainty affects both the quadratic objective function and linear contractual constraints. To solve the resulting problem, we first formulate a deterministic model of the scheduling problem, then establish a min-max robust counterpart, and finally apply some mathematical transformations to solve the equivalent problem. We also deal with the conservatism of the robust control algorithm and flexibility of the method for application to different settings. The validity and effectiveness of the proposed approach is verified by simulation results.

Supply chain disruptions are increasingly caused by growing global supply sourcing, complexity, and interconnectedness of supply chains (SCs). A key challenge in the context of supply chain disruption management is to control and monitor the ripple effect of SCs. The ripple effect occurs when the impact of disruption cannot be localized and propagates throughout the SC. Like the bullwhip effect, the ripple effect can negatively impact performance both upstream and downstream of SC entities. This work proposes a new methodology, based on a probabilistic graphical model, to analyze the ripple effect in a two-stage SC. The probabilistic graphical model developed is capable of capturing disruption propagation that can transfer from upstream suppliers to downstream end customer in an SC.

This paper develops a robust optimization framework for the day-ahead energy scheduling of a grid-connected residential user. The system incorporates a renewable energy source (RES), a battery energy storage system (BESS) as well as elastic controllable and critical noncontrollable electrical appliances. The proposed approach copes with the fluctuation and intermittence of the RES generation and non-controllable load demand by a tractable robust optimization scheme requiring minimum information on the sources of uncertainty. The main objective is minimizing the total energy payment for the user considering operational/technical constraints and a contractual constraint penalizing the excessive use of energy. The presented framework allows the decision maker to define different robustness levels for uncertain variables, and to flexibly establish an equilibrium between user's payment and price of robustness. To validate the effectiveness of the proposed framework under uncertainty, we simulate the dynamics of a residential user as a case study. A comparison between the proposed robust approach and the same method with deterministic RES and loads profiles is carried out and discussed.

This paper proposes a real-time strategy based on Model Predictive Control (MPC) for the energy scheduling of a grid-connected smart residential user equipped with deferrable and non-deferrable electrical appliances, a renewable energy source (RES), and an electrical energy storage system (EESS). The proposed control scheme relies on an iterative finite horizon on-line optimization, implementing a quadratic cost function to minimize the electricity bill of the user's load demand and to limit the peak-to-average ratio (PAR) of the energy consumption profile whilst considering operational constraints. At each time step, the optimization problem is solved providing the cost-optimal energy consumption profile for the user's deferrable loads and the optimal charging/discharging profile for the EESS, taking into account forecast uncertainties by using the most updated predicted values of local RES generation and non-deferrable loads consumption. The performance and effectiveness of the proposed framework are evaluated for a case study where the dynamics of the considered residential energy system is simulated under uncertainties both in the forecast of the RES generation and the non-deferrable loads energy consumption.

Highlights • We presented the first review of application of BN in SC risk and resilience. • We analyzed reviewed journal papers using network analysis and clustering analysis. • The gaps and future research gaps are identified and discussed., In the broad sense, the Bayesian networks (BN) are probabilistic graphical models that possess unique methodical features to model dependencies in complex networks, such as forward and backward propagation (inference) of disruptions. BNs have transitioned from an emerging topic to a growing research area in supply chain (SC) resilience and risk analysis. As a result, there is an acute need to review existing literature to ascertain recent developments and uncover future areas of research. Despite the increasing number of publications on BNs in the domain of SC uncertainty, an extensive review on their application to SC risk and resilience is lacking. To address this gap, we analyzed research articles published in peer-reviewed academic journals from 2007 to 2019 using network analysis, visualization-based scientometric analysis, and clustering analysis. Through this study, we contribute to literature by discussing the challenges of current research, and, more importantly, identifying and proposing future research directions. The results of our survey show that further debate on the theory and application of BNs to SC resilience and risk management is a significant area of interest for both academics and practitioners. The applications of BNs, and their conjunction with machine learning algorithms to solve big data SC problems relating to uncertainty and risk, are also discussed.

Supplier selection is an important strategic decision in the context of supply chain management. Existing literature on the subject of supplier selection is focused on evaluating primary (e.g., cost, quality, lead time) and green (e.g., CO2 emission, environmental practices) criteria. However, the concept of supplier resilience has recently emerged due to advent of competitive and global supply chains (and the operational and disruptive risks to which they are exposed). Several resilience-based supplier selection criteria are developed with respect to absorptive, adaptive, and restorative capacities. This paper further proposes a Bayesian network (BN), a paradigm that effectively models the causal relationships among variables but that has not been used in the context of supplier evaluation and selection, to quantify the appropriateness of suppliers across primary, green, and resilience criteria. Some benefits of the BN paradigm, including an ability to handle expert evidence and to perform sensitivity and propagation analyses, are demonstrated with an initial illustrative example of three suppliers.

The high power dissipation is one of the most important problems of the z-source inverter (ZSI). By using an appropriate optimization scheme, the losses can be significantly reduced without any negative impact on the other characteristics of the inverter. In this paper, a multi-objective optimization is implemented in order to reduce the ZSI total losses as well as to improve the z-source active power filter (APF) performance. The optimization is focused on the four important objectives including power losses of the Z-source APF, the initial cost of the system components, the voltage and current ripples, and the boost factor of the z-source network. For these purposes, the multi-objective genetic algorithm (MOGA) is employed. The numerical and simulation results are presented to evaluate the optimization performance. The results show that a good balance can be achieved between the switching power losses, the voltage-current ripple levels, the component costs and the boost factor using the optimized parameters.

This paper aims to extract a closed-form current-voltage (I–V) equation for short channel carbon nanotube field-effect transistors (CNFETs) in saturation region based on the Stanford model and ballistic relation of one channel CNFET. In order to obtain the goal, firstly, the ballistic relation is simplified to a closed-form I–V equation, and then, it is verified by a linear system identification method based on least square (LS). In addition, an extended quantitative noise analysis is carried out in the circuit level and the results are confirmed in HSPICE based on the CNFET HSPICE model in which the noise sources implemented in Verilog-A are added to the model.

Most of the parameters in meta-heuristic algorithms need precise tuning before applying into the optimization problems. The quality of solutions obtained by meta-heuristic algorithms is highly dependent on how efficiently its parameters are tuned by the users. Setting the parameters is not straightforward, and there is no definitive way to determine their appropriate values other than trial and error, which is an inefficient and time-consuming process. Deviation angle parameter in the imperialist competitive algorithm (ICA) plays an important role as driver of search diversification and intensification. In this paper, we propose a fuzzy adaptive imperialist competitive algorithm (FAICA), where the deviation parameter is adaptively adjusted using a fuzzy controller to optimally control diversification and intensification through the search process. A set of benchmark function tests has been carried out to demonstrate the effectiveness and robustness of the proposed algorithm. The results of experiments are compared with the original ICA and other five well-known algorithms. The experimental results exhibit the merit of achieving balance between intensification and diversification in the search which indicates superiority of FAICA over ICA.

This is the first paper that reviews the application of Imperialist Competitive Algorithm in different engineering disciplines.The development trend of the ICA's applications is analyzed statistically in order to show its popularity.Future research opportunities and directions are discussed to motivate the future researchers. The Imperialist Competitive Algorithm (ICA), derived from the field of human social evolution, is a component of swarm intelligence theory. It was first introduced in 2007 to deal with continuous optimization problems, but recently has been extensively applied to solve discrete optimization problems. This paper reviews the underlying ideas of how ICA emerged and its application to the engineering disciplines mainly on industrial engineering. The present study is the first ever comprehensive review on ICA, which indicates a statistically significant increase in the amount of published research on this metaheuristic algorithm, especially research addressing discrete optimization problems. Future research directions and trends are also described.

Today's manufacturing plants tend to be more flexible due to rapid changes in product mix and market demand. Therefore, this paper investigates the problem of location and relocation (when there are changes incurred to the material flows between departments) manufacturing facilities such that the total cost of material flows and relocation costs are minimized. This problem is known as the dynamic facility layout problem (DFLP), which is a general case of static facility layout problem. This paper proposes a robust and simply structured hybrid technique based on integrating three meta-heuristics: imperialist competitive algorithms, variable neighborhood search, and simulated annealing, to efficiently solve the DFLP. The novel aspect of the proposed algorithm is taking advantage of features of all above three algorithms together. To test the efficiency of our algorithm, a data set from the literature is used for the experimental purpose. The results obtained are quite promising in terms of solution quality for most of the test problems.

Performances of three neural networks, consisting of a multi-layer perceptron, a radial basis function, and a neuro-fuzzy network with local linear model tree training algorithm, in modeling and extracting discriminative features from the response patterns of a temperature-modulated resistive gas sensor are quantitatively compared. For response pattern recording, a voltage staircase containing five steps each with a 20 s plateau is applied to the micro-heater of the sensor, when 12 different target gases, each at 11 concentration levels, are present. In each test, the hidden layer neuron weights are taken as the discriminatory feature vector of the target gas. These vectors are then mapped to a 3D feature space using linear discriminant analysis. The discriminative information content of the feature vectors are determined by the calculation of the Fisher's discriminant ratio, affording quantitative comparison among the success rates achieved by the different neural network structures. The results demonstrate a superior discrimination ratio for features extracted from local linear neuro-fuzzy and radial-basis-function networks with recognition rates of 96.27% and 90.74%, respectively.