Bayesian Belief Network Method Research Articles

LEAN ENERGY EFFICIENCY: SCENARIO ANALYSIS OF A REFRIGERATOR PLANT APPLICATION

Energy concern is increasing in the manufacturing companies implementing the most recent technologies. Energy is a major input for many industries and therefore, within the definition of Industry 4.0 new energy efficiency strategies are defined. Energy flexible processes and waste-to-energy are well-known strategies since they are easily implemented in any manufacturing site. Reduction in energy consumption is also facilitated by the preventive maintenance and renewal of the technologies using energy resources. This study aims to apply a lean production method Value Stream Maps (VSM) on energy consumption levels of the processes to evaluate the decision of technology and/or process change. Processes which use energy excessively when compared with the industry average, will be considered as a bottleneck. Using the value stream maps for energy use is accepted as “Lean Energy Efficiency”. Parameters determined by the preparation of these maps will be enriched by interdependencies determined using the Bayesian Belief Network, which will support finding the priorities among the new efficiency activities. This technique will facilitate the decision of repair or buy through priority scenarios showing the possibility of a decision in each scenario. The combination of the Lean Energy Efficiency Method and the Bayesian Belief Network Method will assist the decision-makers in developing more informed and knowledge-based strategies. The case study is realized in a refrigerator factory handling the energy consumption of different departments. Consequently, these findings could be used as a roadmap for the technology renewal investment decisions made by the firm.

A Bayesian Belief Network method for bridge deterioration detection

Bridges are one of the most important assets of transportation networks. A closure of a bridge can increase the vulnerability of the geographic area served by such networks, as it reduces the number of available routes. Condition monitoring and deterioration detection methods can be used to monitor the health state of a bridge and enable detection of early signs of deterioration. In this paper, a novel Bayesian Belief Network (BBN) methodology for bridge deterioration detection is proposed. A method to build a BBN structure and to define the Conditional Probability Tables (CPTs) is presented first. Then evidence of the bridge behaviour (such as bridge displacement or acceleration due to traffic) is used as an input to the BBN model, the probability of the health state of whole bridge and its elements is updated and the levels of deterioration are detected. The methodology is illustrated using a Finite Element Model (FEM) of a steel truss bridge, and for an in-field post-tensioned concrete bridge.

Applying Bayesian belief networks (BBNs) with stakeholders to explore and codesign options for water resource interventions

Bayesian Belief networks (BBNs) are a useful tool to account for uncertainty and can be used to incorporate stakeholder understandings of how a system works. In this study, BBNs were applied to elicit and discuss local stakeholders’ concerns in conflicts over water resource planning in two cases in southern Thailand. One concerned the construction of a dam proposed by a top-down project. The other concerned a bottom-up participatory process at the catchment scale to assess the need for water resources interventions and explore perceptions on alternative design options. In the top-down project, the responses of participants during the elaboration of the BBN showed that potentially affected stakeholders were particularly concerned about limited consultation and lack of shared benefits, which led them to oppose the dam project. In the bottom-up project, local stakeholders expected and agreed with the benefits of a dam, proposing to locate the dam upstream of community land. The BBN method did not facilitate dialogue in the top-down dam-building project because no alternative design options could be discussed and potentially affected stakeholders did not want to discuss compensation because of mistrust and differences in valuation of effects. In the bottom-up project, the BBN method did facilitate dialogue on alternative intervention options and their effects. The replicable BBN framework can support policy-makers to better understand water conflict situations in different stages of planning. Its application supports exploring a wider repertoire of options, enlarging the scope for more inclusive and sustainable solutions to water resource conflicts.

Analisis Penerima Bantuan Sosial menggunakan Bayesian Belief Network

Bantuan Sosial merupakan pengeluaran berupa uang, barang, atau jasa yang diberikan oleh pemerintah pusat atau daerah kepada masyarakat guna melindungi masyarakat dari kemungkinan terjadinya risiko sosial, meningkatkan kemampuan ekonomi, serta kesejahteraan masyarakat. Pada kesejahteraan masyarakat, terdapat masalah yang masih belum terselesaikan sampai saat ini, yaitu kemiskinan. Kemiskinan merupakan masalah sosial yang masih belum terselesaikan di negara-negara berkembang termasuk diantaranya adalah Indonesia. Pada kemiskinan sendiri, seseorang dinyatakan miskin apabila pendapatanya lebih rendah dari garis kemiskinan serta tidak bisa untuk memenuhi kebutuhan sehari-hari. Dari permasalahan tersebut maka diperlukan analisis lebih lanjut untuk mencari kriteria yang paling berpengaruh yang nantinya dapat digunakan untuk memaksimalkan program yang telah dibuat agar taraf ekonomi pada rumah tangga sasaran dapat bertambah. Penelitian ini dilakukan untuk menganalisis faktor yang mempengaruhi perubahan taraf ekonomi menggunakan metode Bayesian Belief Network. Dari skenario yang telah dilakukan pada pengujian menggunakan metode Bayesian Belief Network ditemukan bahwa peningkatan pada kesejahteraan dan tabungan pada masyarakat di desa Srigading mampu meningkatkan perubahan taraf ekonomi hingga 71%.

Assessing offshore wind turbine reliability and availability

The renewables sector and particularly offshore wind energy is a fast developing industry over the last few years. Especially, activities related to the installation, and operation and maintenance of offshore wind turbines become a challenging task with inherent risks. This article assesses the risks related to the above stages of a wind farm lifecycle using the failure mode, effects and criticality analysis and hazard identification methods. All works, from installation to operation and maintenance, are considered together with the wind turbine main components. An integrated risk analysis methodology is presented addressing personnel Safety (S), Environmental impact (E), Asset integrity (A), and Operation (O). The above is supplemented by a cost analysis with the aid of Bayesian belief networks method to assist the decision-making process related to installation, and operation and maintenance tasks. All major risks and critical wind turbine components are identified as well as measures are suggested to prevent or mitigate them. Moreover, inspection and maintenance plans are elaborated in general for the mentioned activities.

A probabilistic data-driven framework for scoring the preoperative recipient-donor heart transplant survival

Recent research has shown that data mining models can accurately predict the outcome of a heart transplant based on predictors that include patient and donor's health/demographics. These models have not been adopted in practice, however, since they did not: a) consider the interactions between the explanatory variables; b) provide a patient's specific risk of survival (reported results have been primarily deterministic); and c) offer an automated decision tool that can provide some data-driven insights to practitioners. In this study, we attempt to overcome these three limitations through the use of Bayesian Belief Networks (BBN). The proposed BBN framework is comprised of four phases. In the first two phases, the data is preprocessed, and a candidate set of predictors is generated based on employing several variable selection methods. The third phase involves the addition of medically relevant variables to the list. In phase four, the BBN model is applied. The results show that the proposed BBN method provides similar predictive performance to the best approaches in the literature. More importantly, our method provides novel information on the interactions among the predictors and the conditional probability of survival for a given set of relevant donor–recipient characteristics. We offer U.S. practitioners a decision support tool that presents an individualized survival score based on our BBN model (and the UNOS dataset).

Late customization strategy with service levels requirements

Both supply issues and interaction between various stages of a production network are common occurrences in product customization under a delayed differentiation strategy. This paper studies non-decouple systems for product customization in the context of a single-market segment under a delayed differentiation strategy by incorporating supply issues and interaction between various stages of a production network. We analyze two alternatives for customizing two individual products. One alternative is the contractual obligations for raw materials delivery where the supplier is responsible for maintaining certain inventory service levels agreements. The other alternative is the customer demand requirements where the manufacturer is responsible for maintaining certain end-products service levels to prevent customer erosion. We characterize the service level requirements in both scenarios and determine the optimal customization point of the production network where the practitioner can operate at minimum cost. We use a Bayesian Belief Networks method to model interaction between stages of the production network and derive the inventory service level requirements. A mini-case involving the customization of a personal desktop computer is used to illustrate the applicability of this framework. We also compare the benefits of delayed product differentiation under non-decouple systems and the traditional decouple systems and provide insights into how firms can choose the right strategy to effectively compete.

Life-cycle Risk Modeling: Alternate Methods Using Bayesian Belief Networks

In recent years natural and man-made disasters have highlighted the need for robust supply chain risk management (SCRM) in manufacturing firms from a life-cycle perspective (pre-manufacturing, manufacturing, use, post-use stages). Bayesian Belief Networks (BBN) provide a means to probabilistically represent risk interdependencies and to proactively identify and manage any existing vulnerabilities. In this work, the BBN method is implemented for a product in the aerospace industry. Risk network maps are developed to identify interdependencies and describe the potential risk propagation behavior during each life-cycle phase and from one phase to another. Due to limited number of respondents and lack of certainty with respect to the post-use phase, enhanced methods of risk likelihood assessment are necessary specifically for the post-use phase assessment. In this paper two alternate techniques are compared for risk modeling using BBN in such situations: Boolean nodes and numeric simulation nodes. Results show that numeric nodes provide a more thorough explanation of the interconnections of the risk items modeled. Further enhancement using an approach that combines both BBN and System Dynamics (SD) for SCRM is discussed and possible variations for linking variables between SD and BBN are also presented.

A method of sensor fault detection and identification

A method of Bayesian belief network (BBN)-based sensor fault detection and identification is presented. It is applicable to processes operating in transient or at steady-state. A single-sensor BBN model with adaptable nodes is used to handle cases in which process is in transient. The single-sensor BBN model is used as a building block to develop a multi-stage BBN model for all sensors in the process under consideration. In the context of BBN, conditional probability data represents correlation between process measurable variables. For a multi-stage BBN model, the conditional probability data should be available at each time instant during transient periods. This requires generating and processing a massive data bank that reduces computational efficiency. This paper presents a method that reduces the size of the required conditional probability data to one set. The method improves the computational efficiency without sacrificing detection and identification effectiveness. It is applicable to model- and data-driven techniques of generating conditional probability data. Therefore, there is no limitation on the source of process information. Through real-time operation and simulation of two processes, the application and performance of the proposed BBN method are shown. Detection and identification of different sensor fault types (bias, drift and noise) are presented. For one process, a first-principles model is used to generate the conditional probability data, while for the other, real-time process data (measurements) are used.