Asset Condition Research Articles

Enhanced Dual Convolutional Neural Network Model Using Explainable Artificial Intelligence of Fault Prioritization for Industrial 4.0.

Artificial intelligence (AI) systems are increasingly used in corporate security measures to predict the status of assets and suggest appropriate procedures. These programs are also designed to reduce repair time. One way to create an efficient system is to integrate physical repair agents with a computerized management system to develop an intelligent system. To address this, there is a need for a new technique to assist operators in interacting with a predictive system using natural language. The system also uses double neural network convolutional models to analyze device data. For fault prioritization, a technique utilizing fuzzy logic is presented. This strategy ranks the flaws based on the harm or expense they produce. However, the method's success relies on ongoing improvement in spoken language comprehension through language modification and query processing. To carry out this technique, a conversation-driven design is necessary. This type of learning relies on actual experiences with the assistants to provide efficient learning data for language and interaction models. These models can be trained to have more natural conversations. To improve accuracy, academics should construct and maintain publicly usable training sets to update word vectors. We proposed the model dataset (DS) with the Adam (AD) optimizer, Ridge Regression (RR) and Feature Mapping (FP). Our proposed algorithm has been coined with an appropriate acronym DSADRRFP. The same proposed approach aims to leverage each component's benefits to enhance the predictive model's overall performance and precision. This ensures the model is up-to-date and accurate. In conclusion, an AI system integrated with physical repair agents is a useful tool in corporate security measures. However, it needs to be refined to extract data from the operating system and to interact with users in a natural language. The system also needs to be constantly updated to improve accuracy.

Priority Handling of Teurebeh Irrigation Assets, Aceh Besar District, Aceh Province

Teurebeh Irrigation scheme is one of the promising irrigation areas in Aceh Besar District, Aceh Province. It is also one of the buffers in the agricultural sector, covering 750 hectares of agricultural area. However, the construction has shrunk over time, and the evaluation at the end of 2020 classified it as "Poor Performance and Requiring Attention." This study aims to determine the handling priority for the irrigation networks at the Teurebeh Irrigation scheme. Data on irrigation network inventory was used as input. The determination of handling priority was conducted using the Mann-Whitney method. The ranking for determining the importance of irrigation assets is based on the condition and function of the irrigation assets. ArcGIS and Google Earth software were used to determine the condition and handling of all assets. The results show that the handling priority levels are between the primary and secondary channels. The tapping structure of BDT.8, BDT.6, BDT.4, BDT.1, and BDT.7 are ranked 1 to 5, meaning they must be fixed immediately as the conditions are severe to moderately damaged and not functioning to functioning poorly. BDT.1 to BDT.3 is in the Primary Channel, while BDT.4 to BDT.8 is in the Secondary Channel. The handling priority for irrigation construction in the primary channel (BDT.1, BDT.1a, and BDT.1c) and the secondary channel (BDT.8, BDT.6, and BDT.7a) are within priority 1 to 3. The Mann-Whitney test on the proposed hypothesis comparing the urgency level between the two channels is insignificant, indicating no significant difference between the handling of primary and secondary channels, as both have similar conditions and functions. The results are expected to contribute to decision-making at the Irrigation Department in designing proposals for renovating irrigation networks.

Implementation of asset condition models at E-REDES: what comes next?

Implementation of asset condition models at E-REDES: what comes next?

Data-Driven Asset Health Index – an application to evaluate Quay Cranes in container ports

ABSTRACT Ports worldwide seek for more efficient and sustainable operations, relying on critical and valuable assets as their cranes. Internet of Things (IoT) technologies enable a better understanding of critical assets’ behaviour and lifecycle based on data, and development of asset management decision support tools. The Asset Health Index (AHI) is an asset condition score, designed to characterise the state of complex assets. This paper describes the process of developing a data-driven AHI. IoT sensors and Programmable Logic Controllers (PLC) in real operation time, together with information systems, and other more traditional sources of data are some of the data sources for the AHI. The index captures and quantifies dimensions affecting the asset lifecycle, allowing the comparison of the condition of assets and prioritise maintenance investments. The case study is conducted in Quay Cranes of the Port of Felixstowe (UK) as the object of this study and proof of concept.

가상자산에 대한 소득세 부과 방안 연구

Looking at the legal status of virtual assets in Korea, the Tax Act defines “virtual asset” as an electronic certificate that has economic value and can be transacted or transferred electronically. The Tax Act defines “virtual asset” as an electronic certificate (including any rights related to it) that has economic value and can be transacted or transferred electronically. ) that can be transacted or transferred electronically as a thing with economic value, and whether it falls under the civil law is divided into positive and negative theories. The Supreme Court The Supreme Court considers it to be property under the Criminal Code, but not currency under the Bank of Korea Act. Gold under the Capital Market Act does not fall under the definition of a financial investment product, but it is necessary to expand the concept to provide legal safeguards for investor protection. to protect investors.
 E-finance transactions through blockchain are decentralized, meaning that transaction details are shared and stored among all network participants, and all transactions are notarized and managed by participants through transaction confirmation, so there is no centralized administrator. Under the current income tax law, the taxation structure for virtual asset income is based on direct transactions (P2P) that do not go through virtual asset operators, and since it is difficult to identify international transactions, various types of distribution taxes are imposed for domestic transactions, but we propose to approach international transactions by setting international standards and establishing an information exchange system between countries.
 To identify domestic transactions of virtual assets, virtual assets are generally classified as securities or not, and securities-type virtual assets are considered as 'securities' because they meet the 'Howey test', so it is possible to secure transaction details and manage tax revenue through securities transaction taxation. Except for 'securities-type virtual assets', virtual assets are subject to VAT because they are subject to asset exchange transactions, and if they are distributed from Korea to foreign countries, they can be applied at a zero-rated rate as domestic goods. Since the “electronic wallet” of virtual assets refers to the virtual assets themselves, we would like to propose a method of imposing stamp tax on example savings transaction contracts and certificates based on virtual assets, and a method of imposing registration license tax through the virtual asset electronic register by establishing a study managed by the central government. Sanctions for unregistered transactions of virtual assets and a trial period before the introduction of the system will be implemented to encourage positive transactions.
 To identify international transactions of virtual assets, it is necessary to establish a virtual asset transaction information exchange system between countries and agree on standards for international taxation targets and methods to exercise taxation rights in an information cooperation council.

Real-Time Humidity Monitoring Using Distributed Optical Sensor for Water Asset Condition Assessment

It is well accepted that moisture ingress in concrete reduces durability and life span of water assets. Condition assessment is an important tool to inform decision for maintenance, retrofit or replacement. However, the most significant challenge is to obtain accurate condition information, particularly when the inspection points are physically difficult to access or inaccessible. Therefore, a reliable and cost-effective monitoring (sensor) system, preferably real-time with ability to streaming online, would be a useful management tool, particularly for water utilities. This paper describes an approach to develop a distributed optical fibre humidly sensor for condition assessment and environmental monitoring both inside and outside of infrastructures, such as inside the concrete and surrounding soil. A new polyelectrolyte multilayer (PEM) coating with higher sensitive was evaluated for relative humidity measurement in soil and concrete, respectively. In this study, two simulated conditions, in concrete and soil, were conducted to evaluate the sensing concept with the development of appropriate measuring methodologies including fibre installation and protection. The optical fibre sensor setup in laboratory environment showed that optical sensor can detect and indicate voltage change with the variation of moisture contents in both soil and concrete. The test results indicate a good correlation between high levels of relative humidity/moisture and transmitted optical power. A simple relative humidity (RH) calibration can be used to convert signal to RH in percentage for soil and concrete measurements and the procedure used to imbed the fibre in both samples is effective. Nevertheless, the sensor measures soil humidity (not moisture content); therefore, further investigation is required to identify the consequence for the variation of the measured parameter.

Livelihood strategies of farmers in Loksado District, Hulu Sungai Selatan Regency, Indonesia

This research was conducted in Loksado District, Hulu Sungai Selatan Regency, South Kalimantan, Indonesia. Demographic changes and land use changes that occurred in Loksado District resulted in a decrease in the area of farmland. This condition gave rise to changes in the conditions of field farmers in meeting household needs. This study aimed to analyze the livelihood strategies farm farmer households in Loksado District applied. This research is based on the application of qualitative research. Informants in this study were shifting cultivation farmers from the Dayak Meratus Tribe in Haratai Village, Loksado Village, and Loklahung Village, Loksado District, which was selected by snowball sampling. The results of this study indicate that there are three livelihood strategies for farm household farmers: accumulation, consolidation, and survival. The dominant strategy base is agricultural diversification to meet household needs. The selection of household livelihood strategies is related to the socio-economic conditions of the household, the use of resources (the condition of assets owned, domestic workers of working age), and job diversification.

Latent indicators for temporal-preserving latent variable models in vibration-based condition monitoring under non-stationary conditions

Condition-based monitoring for critical assets is reliant on the quality of the indicators used for condition inference. These indicators must be sensitive to the development of faults under constant and non-stationary operating conditions. Latent variable models in the time-preserving framework offer a powerful learning-based technique for the monitoring of critical assets as they only require healthy asset data, and provide indicators from the data space for asset condition inference. However, the latent manifold of latent variable models is often disregarded in favour of data space indicators, such as reconstruction errors, which are primarily focused on measuring the likelihood of the observed data. The latent space is often unintentionally unutilised. In this work, we highlight that the latent manifold is a powerful resource for condition inference and should be utilised for condition monitoring. We conceptually categorise and identify five classes of latent space health indicators that capture various manifold perspectives. These five classes introduce and allow for latent health indicator derivation and are useful for the condition inference task. Fifteen latent health indicators are considered in this work and are applied to the fault diagnostics task on two experimental datasets. An ensemble-based inference procedure is used, which produces a modular fault diagnosis framework. The indicators are shown to be informative for condition inference in both constant and variable operating conditions. Utilising the latent manifold in condition monitoring tasks is important to further develop the learning-based condition monitoring field.

Assessment of Level of Service for Roads under Performance Based Road Maintenance in Kenya

Purpose: This paper seeks to assess Level of Service (LOS) of roads under performance-based road maintenance (PBRM) in Kenya. This was done through a comparison of actual ratings with performance targets defined by the agencies for each asset item within a sample unit. The comparison was used to determine if each asset item had been preserved at minimum acceptable Level of Service (LOS) with objective of establishing if there are any benefits accrued from adopting the PBRM method.
 Methodology: Seven roads were identified in the study area and were divided into 14 sample units, each sample unit containing different asset items to be assessed. Data from the field was collected and recorded using field measurement tools for linear measurements. Dynamic Response Intelligent equipment (DRIMS) was used for measurement of International Roughness Index (IRI). Raw data was converted into performance information using guidelines from Quality Assurance in Highway Maintenance Programs and measurement of conformance conducted using Schedule Performance Index.
 Findings: The performance obtained indicated that the goal objective for each road was not achieved with an overall performance deviating from the target performance by 25.05%. However, there was 10% compliance at the asset group level and 35% compliance at asset item level. Noncompliance at road section level, 10% compliance at asset group level and 35% compliance at asset item level signifies inadequacies in PBRM practice in Kenya. This informs need for improvement by road agencies and contractors for optimum benefits from PBRM. There is need for contractor capacity building and enhanced partnering environment between contractors and management engineering consultants to provide technical assistance for improved contractor compliance in PBRM practice.
 Unique Contribution to Theory, Practice and Policy: The Concept of Performance based road maintenance (PBRM) is meant to solve problems arising from traditional contracting methods. As demonstrated in this study, effective application of PBRM principles results in cost savings in managing and maintaining roads, expenditure certainty, reduced inhouse workforce, greater road user satisfaction and improved conditions of contracted road assets. The requirement for contractors to provide a product of predefined quality is a definite means of ensuring that the road asset is performing over extended time. It is recommended that PBRM is adopted widely in road asset management to achieve an effective and efficient road network.

Study Physiology and Immunology Emergency Clinic Diseases in Patients Experiencing Clinical Consideration and Contamination.

Emergency clinic diseases or medicinal services related contaminations happen in patients experiencing clinical consideration. This contamination happens worldwide in both created and creating nations. Generally, Staph, Streptococcus, an d Escherichia coli have been triple clinic diseases. Emergency clinic pneumonia, careful injury contaminations, and vascular-related septicemia made access most illnesses and passings in medical clinic patients; concentrated consideration units were the foca l point of anti-infection obstruction. The procured antimicrobial opposition is the primary issue, and the obstruction of Staphylococcus aureus is the greatest concern. The move to outpatient care departs the most powerless patients in medical clinics, with emergency clinic disease representing 7% in created nations and 10% in creating nations. Since this disease happens duri ng medical clinic remain, it causes delayed remain, incapacity and financial weight. Visit diseases incorporate focal line circulato ry system contaminations, urinary tract diseases related with a catheter, careful site contaminations and ventilation-related pneumonia. Clinic pathogens incorporate microscopic organisms, infections, and contagious parasites. As indicated by WHO gauges, almost 15% of all patients endure in medical clinics these contaminations. During hospitalization, the patient is presented to pathogens through an alternate asset condition, human services work force, and other contaminated patients.

Influence of an integrated value-based asset condition assessment in built asset management

Purpose Built asset management processes require a long transition period to collect, edit and update asset conditions information from existing data sets. This paper aims to explore and explain whether and how digital technologies, including asset information model (AIM), Internet of Things (IoT) and blockchain, can enhance asset conditions assessment and lead to better asset management. Design/methodology/approach Mixed methods are applied to achieve the research objective with a focus in universities. The questionnaire aims to test whether the integration of AIM, IoT and blockchain can enhance asset condition assessment (ACA). Descriptive statistical analysis was applied to the quantitative data. The mean, median, mode, standard deviation, variance, skewness and range of the data group were calculated. Semi-structured interviews were designed to answer how the integration of AIM, IoT and blockchain can enhance the ACA. Quantitative data was analysed to define and explain the essential factors for each sub-hypothesis. Meanwhile, to strengthen the evaluation of the research hypothesis, the researcher also obtained secondary data from the literature review. Findings The research shows that the integration of AIM, IoT and blockchain strongly influences asset conditions assessment. The integration of AIM, IoT and blockchain can improve the asset monitoring and diagnostics through its life cycle and in different aspects, including financial, physical, functional and sustainability. Moreover, the integration of AIM, IoT and blockchain can enhance cross-functional collaboration to avoid misunderstandings, various barriers and enhance trust, communication and collaboration between the team members. Finally, costs and risk could be reduced, and performance could be increased during the ACA. Practical implications The contribution of this study indicated that the integration of AIM, IoT and blockchain application in asset assessment could increase the efficiency, accuracy, stability and flexibility of asset assessment to ensure the reliability of assets and lead to a high-efficiency working environment. More importantly, a key performance indicator for ACA based on the asset information, technology and people experience could be developed gradually. Originality/value This study can break the gap between transdisciplinary knowledge to improve the integration of people, technology (AIM, IoT and blockchain) and process value-based ACA in built asset management within universities.

Integrating condition-based monitoring with process sensors information for operation and maintenance in a functional modeling framework

Safe operations and adequate maintenance are two main means to achieve reliable production and reduce downtime of a plant. While the tasks of operations and maintenance are carried out by two different groups of staff, as a result, the close relationship between the two tasks is split. In this paper, this challenge is handled by a proposed integrated functional modeling framework. In this framework, the Multilevel Flow Modeling (MFM) method with its cause-consequence reasoning rules is used. Condition-based monitoring is a well-accepted strategy for predictive maintenance and fault detection based on measurements is a well-developed technology for operation support. Information fusion including monitoring conditions of assets and process sensors information for both operation and maintenance in the same modeling framework is desired. The qualitative relationship distribution between operations and maintenance can be established based on the function states of the system. In addition, these relationships are visible for both groups of staff. As a result, the detected information in the early stage of the development of the unpleasant scenarios is used to improve their situation awareness, so that the undesired emergency shutdown from both perspectives of operation and maintenance is prevented. Consequently, it can reduce production loss. A case study of operations and maintenance of a seawater injection system is carried out and shows the industrial applicability of the proposed framework. The case study strongly reveals that there is a highly close relation between operation and maintenance for ensuring the system working properly. It demonstrates that the proposed integrated framework is not only able to support operational tasks but also for the maintenance tasks by including relevant maintenance information of the system. The results show that it can potentially help with decreasing downtime of the system.

Features of Project Management of Industrial Enterprise Reconstruction on the Basis of Technical, Economic, Legal and Technological Factors

One of the main forms of industrial enterprise development is reconstruction. The main task is to improve the technical level of the enterprise, the quality of products, introduce energy-saving and resource-saving technologies using existing fixed assets. There is a growing interest in reducing maintenance and repair costs, as well as in significantly increasing the time between repairs and reducing the energy intensity of buildings. This article is devoted to the peculiarities of managing a project for the reconstruction of an industrial enterprise, which includes a cultural heritage site. The study was conducted on the example of one of the oldest enterprises in Russia - Joint Stock Company 'Votkinsky Zavod'. The purpose of this study is to develop a methodological basis for the effective management of the reconstruction of an industrial facility based on a database of the technical condition of the main production assets of the Joint Stock Company 'Votkinsky Zavod'. In the course of the study, a detailed analysis of the historically established fund of buildings and structures, in-cluding a cultural heritage site, was carried out with the aim of their possible most optimal further adaptation. The main periods of the development of the plant in the history of its existence are highlighted. The result of the study is an assessment of the technical condition of buildings and structures of the industrial en-terprise under consideration, the classification of a group of buildings according to the periods of construction, according to design schemes and materials used, as well as according to the main defects and damage to structures. As a result of the study, a list of the most common defects and damages to the building structures of buildings and structures was compiled, and a base of technical information on the state of the enterprise's fixed production assets was formed. An algorithm for choosing the optimal variant of the project for the reconstruction of a building, structure or complex is proposed based on the observance of a certain sequence of economic analysis when considering technical, economic, social, aesthetic and time indicators with comparability of design solutions for reconstruction options.

Optimizing Predictive Maintenance Decisions: Use of Non-Arbitrary Multi-Covariate Bands in a Novel Condition Assessment under a Machine Learning Approach

Jointing Condition-Based Maintenance (CBM) with the Proportional Hazards Model (PHM), asset-intensive industries often monitor vital covariates to predict failure rate, the reliability function, and maintenance decisions. This analysis requires defining the transition probabilities of asset conditions evolving among states over time. When only one covariate is assessed, the model’s parameters are commonly obtained from expert opinions to provide state bands directly. However, the challenge lies within multiple covariate problems, where arbitrary judgment can be difficult and debatable, since the composite measurement does not represent any physical magnitude. In addition, selecting covariates lacks procedures to prioritize the most relevant ones. Therefore, the present work aimed to determine multiple covariate bands for the transition probability matrix via supervised classification and unsupervised clustering. We used Machine Learning (ML) to strengthen the PHM model and to complement expert knowledge. This paper allows obtaining the number of covariate bands and the optimal limits of each one when dealing with predictive maintenance decisions. This novel proposal of an ML condition assessment is a robust alternative to the expert criterion to provide accurate results, increasing the expectation of the remaining useful life for critical assets. Finally, this research has built an enriched bridge between the decision areas of predictive maintenance and Data Science.

Application Research of Portfolio Related Theory--Based on Hong Kong Stock Data

With the development of the stock market, people have carried out in-depth and comprehensive research on the portfolio theory. This paper aims to verify and analyze the practical application of the relevant theories. By collecting the stock price information of Hong Kong stocks as the target, this paper verifies and discusses basic theories such as tangent portfolio theory, portfolio diversification effect, Markowitz portfolio theory and Sharpe ratio. In the minimum variance portfolio and the tangent portfolio, this paper obtained the weight of each asset under the conditions of assets and without risk-free assets. Finally, the differences in the weights of investment targets in the portfolio are analyzed and discussed. The results show that, first, in the minimum variance portfolio, when considering risk-free assets, only investing in risk-free assets is considered, and when risk-free assets are not considered, assets with lower variance are given higher weights. Second, in a tangent portfolio, regardless of whether risk-free assets are considered, assets with higher Sharpe ratios in the portfolio are given higher weights. This can give some guidance and suggestions for future investment behavior.

SOSIALISASI TATA CARA PENDAFTARAN TANAH WAKAF DI MUI KABUPATEN SIMALUNGUN

Waqf institutions are one of the Islamic social institutions that have high socio-economic value, and are a form of embodiment of social justice in Islam. in general, so it is necessary to develop its utilization in accordance with sharia principles, as mentioned in the Al-Qur'an surah At-Taubah verse 103. However, in reality many waqf lands are neglected in the community, giving rise to legal problems, especially in Simalungun Regency. This is due to Nazhir's inability to manage and develop waqf assets, the attitude of the people who do not care, the community does not understand the status of waqf assets that should be protected, and the community does not understand how to register waqf land. This problem encourages extension agents to cooperate with Ladui MUI Province, MUI Simalungun Regency, to conduct legal counseling entitled Procedures for Registration of Waqf Land in the form of Community Service at the Simalungun Regency MUI Office. Ways and conditions for registering waqf land, understanding the elements of waqf, nazhir waqf obligations, understanding legal issues regarding waqf and the legal force of waqf certificates. The conclusion in this activity is that it is necessary to increase the frequency and socialization regarding. Procedures for Waqf Land Registration.

Using image analysis to quantify defects and prioritize repairs in built-up roofs

PurposeThis paper aims to use convolutional neural networks (CNNs) to provide an objective approach to classify deteriorated building assets according to the type and extent of damage. This research supports automated inspection of buildings and focuses on roofing elements as one of the most critical and externally distressed elements in buildings.Design/methodology/approachIn this paper, 5,000+ images of deteriorated roofs from several buildings were collected to design a CNN system that automatically identifies and sizes roofing defects. Experimenting with different CNN formulations, the best accuracy is achieved using two-stage CNNs. The first-stage CNN classifies images into defect/no defect, while the second stage classifies the defected images according to the damage type. Based on the image classification, optimization is used to prioritize roof repairs by maximizing the return from limited rehabilitation funds.FindingsThe developed CNNs reached 95% and 97% accuracy for the first and second phases, respectively, which is higher than achieved in previous literature efforts. Using the proposed model to automate inspection and condition assessment activities proved to be faster than conventional methods. Repair/replace strategy for a case study of 21 campus buildings based on their condition and budgetary constraints was suggested.Research limitations/implicationsFuture research includes testing different data acquisition technologies (e.g. infrared imaging), performing severity-based classification and integrating with BIM for defect localization.Originality/valueThis study provides an objective approach to automate asset condition assessment and improve funding decisions using a combination of image analysis and optimization techniques. The proposed approach is applicable toward other asset types and components.

Risk-based asset life management for steam turbines and generators

ABSTRACT To ensure a good return on investment, it is important for power plant operators to set up an asset management program that will ensure reliable operation over the commercial life of the plant. General Electric (GE) has been applying a variety of asset management technologies to improve power plant life and performance for our customers. These methods consider various inputs including equipment design knowledge; asset condition assessed from real-time monitoring and inspection data; and lessons learned from repairing, operating and maintaining thousands of assets globally. In this paper, a subset of that information is applied to develop a methodology to perform risk-based health and life assessments for steam turbines and generators. The method has been used on several steam power plants around the world, to help operators of ageing steam plants make the right decisions related to end-of-life management or life extension.

Street closure prediction based on the combined conditions of spatially collocated municipal infrastructure assets at the segment level

Generally, intervention activities that require excavations tend to create negative socio-economic impacts, such as increased traffic congestion and travel time, noise and air pollution, etc. A holistic form of intervention planning is needed to minimize the negative impacts and the frequency of maintenance, repair, and rehabilitation activities resulting in partial or complete street closures. This planning should be based on a unified classification model, which evaluates the conditions of different municipal assets (i.e., water and sewer pipes and road pavements) at the segment-level. The objectives of this paper are: (1) Developing a machine learning method for systematic condition classification of different spatially collocated underground municipal assets within a segment; and (2) Applying a heuristic approach for determining street closures based on the synchronized or unsynchronized interventions at the segment level induced from combining the interventions of individual assets within each segment. The conditions of all three assets are determined using three independent ensemble machine learning models, each made up of Random Forest, Gradient Boosted Trees, and Deep Learning as base learners. The required asset-level interventions are determined by these conditions combined with heuristics specific to each spatially collocated asset in a street segment to decide the segment-level synchronized or unsynchronized interventions and the nature of the street closures (partial, complete, or both) needed to undertake the interventions. The accuracies of the three models are 98.64%, 96.37%, and 82.38% for the pavement, water and sewer pipes, respectively. The segment-level intervention strategies resulted in both synchronized and unsynchronized interventions leading to complete, partial, or a combination of both partial and complete street closures. The predicted segment-level interventions had an accuracy of 79.92%. This research will aid municipal decision-makers in identifying and prioritizing synchronized segment-level interventions, therefore, improving the planning and estimation of the intervention durations and street closures.

Automatic Detection of Corrosion in Large-Scale Industrial Buildings Based on Artificial Intelligence and Unmanned Aerial Vehicles

In recent years, Artificial Intelligence (AI) provided essential tools to enhance the productivity of activities related to civil engineering, particularly in design, construction, and maintenance. In this framework, the present work proposes a novel AI computer vision methodology for automatically identifying the corrosion phenomenon on roofing systems of large-scale industrial buildings. The proposed method can be incorporated into computational packages for easier integration by the industry to enhance the inspection activities’ performance. For this purpose, a dedicated image database with more than 8k high-resolution aerial images was developed for supervised training. An Unmanned Aerial Vehicle (UAV) was used to acquire remote georeferenced images safely and efficiently. The corrosion anomalies were manually annotated using a segmentation strategy summing up 18,381 instances. These anomalies were identified through instance segmentation using the Mask based Region-Convolution Neural Network (Mask R-CNN) framework adjusted to the created dataset. Some adjustments were performed to enhance the performance of the classification model, particularly defining an adequate input image size, data augmentation strategy, Intersection over a Union (IoU) threshold during training, and type of backbone network. The inferences show promising results, with correct detections even under complex backgrounds, poor illumination conditions, and instances of significantly reduced dimensions. Furthermore, in scenarios without a roofing system, the model proved reliable, not producing any false positive occurrences. The best model achieved metrics’ values equal to 65.1% for the bounding box detection Average Precision (AP) and 59.2% for the mask AP, considering an IoU of 50%. Regarding classification metrics, the precision and recall were equal to 85.8% and 84.0%, respectively. The developed methodology proved to be extremely valuable for guiding infrastructure managers in taking physically informed decisions based on the real assets condition.