•A generic solution approach applicable to discrete types of infrastructure assets.•Development of predictive models for efficient maintenance of railway switches.•Utilize data from an in-use ...business process from a real railway agency.•Explanation of the models’ performance by feature and instance level interpretation.
With growing service demands, rapid deterioration due to extensive usage, and limited maintenance due to budget cuts, the railway infrastructure is in a critical state and require continuous maintenance. The infrastructure managers have to come up with smart maintenance decisions in order to improve the assets’ condition, spend optimal cost and keep the network available. Currently, the infrastructure managers lack the tools and decision support models that could assist them in taking (un) planned maintenance decisions effectively and efficiently. Recently, many literature studies have proposed to employ the machine learning techniques to estimate the performance state of an asset, predict the maintenance need, possible failure modes, and such similar aspects in advance. Most of these studies have utilised additional data collection measures to record the assets’ behaviour. Though useful for experimentation, it is expensive and impractical to mount monitoring devices on multiple assets across the network. Therefore, the objective of this study is to develop predictive models that utilise existing data from a railway agency and yield interpretable results. We propose to leverage the tree-based classification techniques of machine learning in order to predict maintenance need, activity type and trigger’s status of railway switches. Using the data from an in-use business process, predictive models based on the decision tree, random forest, and gradient boosted trees are developed. Moreover, to facilitate in models interpretability, we provided a detail explanation of models’ predictions by features importance analysis and instance level details. Our solution approach of predictive models development and their results explanation have wider applicability and can be used for other asset types and different (maintenance) planning scenarios.
Data-driven decision support can substantially aid in smart and efficient maintenance planning of road bridges. However, many infrastructure managers primary rely on information obtained during ...visual inspection to subjectively decide on the follow-up maintenance actions. The subjective approach is likely to lack the appropriate use of inspection data and does not promise cost-effective maintenance plans. In this paper, we show that the historical and operational data, readily available at the agencies, is of vital importance and can be used effectively for the recommendations of maintenance advises for bridges. This is achieved by developing a machine learning system that is trained on the past asset management data and provide support to the decision-makers in the condition assessment, risk analysis, and maintenance planning tasks. We have evaluated several traditional learning algorithms as well as the deep neural networks with entity embedding to find the optimal predictive models in terms of predictive capability. Additionally, we have explored the multi-task learning framework that has a shared representation of related prediction tasks to develop a powerful unified model. The analysis of results shows that a unified multi-task learning model performed best for the considered problems followed by task-specific neural networks with entity embedding and class weights. The results of models are further evaluated by instance-level explanations, which provide insights about essential features and explain the importance of data attributes for a particular task.
•Predictive modeling of maintenance related tasks (interventions) of bridges using historical data•Employing entity embedding with neural networks for structured categorical data•Using multi-task learning to learn shared representations of related predictive tasks•Providing instance-level interpretation of results of the models
Display omitted
•Bridging the gap between sewer condition data-needs and suitable inspection methods.•Categorizing sewer inspection methods based on their data output.•Identifying required research ...areas for data-driven sewer asset management.•Improving structural and environmental inspection methods is a future research need.•Enhancing sewer assessment data quality and its automation requires a holistic inspection approach.
Data-driven sewer asset management uses digital sewer representations to store inspection data and to support predictive maintenance planning. This approach requires asset managers to determine what inspection data they need to collect for the assessment of the asset conditions. Existing studies review sewer inspection methods based on their technical working principles but do not explicitly address what data about condition cues these methods provide. Consequently, literature lacks structured insights that help sewer asset managers link their data-needs with appropriate condition assessment methods. To make this link, we propose a data-needs based categorization of sewer inspection methods. Specifically, we relate data output of inspection methods to condition cues using the classification of hydraulic, structural, and environmental inspection domains. This shows that few methods exist to collect data about cues in structural and environmental domains. Future research should develop methods to satisfy these needs, and eventually, contribute to holistic data-driven asset management.
This paper introduces a comprehensive framework for the development of optimal multi-year maintenance plans for a large number of bridges. A maintenance plan is said to be optimal when, within the ...given budget, a maximum number of bridges can be maintained in the best possible year, achieving maximum performance with minimum socio-economic impact. The framework incorporates heuristic rules, multi-attribute utility theory, discrete Markov chain process, and genetic algorithms to find an optimal balance between limited budgets and performance requirements. The applicability of the proposed framework is illustrated on an extensive case study of highway bridges. The framework enables asset owners to execute various planning scenarios under different budget and performance requirements, where each resulting plan is optimal. The focus of this study has mainly been on highway bridges, however the framework is general and can be applied to any other infrastructure asset type.
The monitoring and tracking of construction equipment, e.g., excavators, is of great interest to improve the productivity, safety, and sustainability of construction projects. In recent years, ...digital technologies are leveraged to develop monitoring systems for construction equipment. These systems are commonly used to detect and/or track different pieces of equipment. However, the recent research work has indicated that the performance of the equipment monitoring system improves when they are able to also recognize/track the activities of the equipment (e.g., digging, compacting, etc.). Nevertheless, the current direction of research on equipment activity recognition is gravitating towards the use of deep learning methods. While very promising, the performance of deep learning methods is predicated on the comprehensiveness of the dataset used for training the model. Given the wide variations of construction equipment, in size and shape, the development of a comprehensive dataset can be challenging. This research hypothesizes that through the use of a robust feature augmentation method, shallow models, such as Random Forest, can yield a comparable performance without requiring a large and comprehensive dataset. Therefore, this research proposes a novel machine learning method based on the integration of Random Forest classifier with the fractional calculus-based feature augmentation technique to develop an accurate activity recognition model using a limited dataset. This method is implemented and applied to three case studies. In the first case study, the operations of two different models of excavators (one small-size and one medium-size) were tracked. By using the data from one excavator for the training and the data from the other one for testing, the impact of equipment size and operators' skill level on the performance of the proposed method is investigated. In the second case study, the data from an actual excavator was used to predict the activity of a scaled remotely controlled excavator. In the last case study, the proposed method was applied for rollers (as an example of non-articulating equipment). It is shown that the fractional feature augmentation method can have a positive impact on the performance of all machine learning methods studied in this research (i.e., Neural Network and Support Vector Machine). It is also shown that the proposed Fractional Random Forest method is able to provide comparable results to deep learning methods using considerably smaller training dataset.
•A novel method for the activity recognition of construction equipment is presented.•The method is based on the integration of Fractional Calculus and Random Forest.•Several case studies are presented to demonstrate the performance of the method.•The augmentation of feature domain by fractional calculus is shown to improve the performance.•Fractional random forest reduces the dependency of activity recognition models on large datasets.
Virtual Reality (VR) based training simulators are successfully employed in many industries (e.g., aviation) to help train operators and professionals in a safe environment. The construction industry ...has also started to use this technology in recent years for training operators of heavy equipment. However, the context presented in the available training simulators is unrealistic because in many instances the training takes place in static sites where there is no mobility in the site. To realistically introduce the context of construction sites into VR scenes sensory data from actual projects can be used. However, currently, there is no systematic insight into (1) the dimensions of context that need to be present in a training simulator, (2) the types of data required to represent various dimensions of the context, and (3) methods for converting context data into a coherent context-realistic training scene that enables bidirectional feedback between trainees and the VR scene. Therefore, this research aims to develop a novel framework to generate coherent context-realistic training simulators from data collected from actual construction projects to enhance construction training simulators. The proposed framework provides a step-wise guideline into (a) collection of appropriate data for context-realistic simulators, (b) development of agents and simulation physics from actual site data and their integration into a scene, (c) scene-trainee interactions in context-realistic scenes, and (d) context-based assessment of the trainees' performance from safety, productivity, and quality perspectives. A prototype is developed and a case study is conducted to demonstrate the feasibility of the proposed framework. A workshop with expert training instructors is conducted to evaluate the effectiveness of the proposed framework for improving simulator-based training. It is shown that compared to the existing simulators, the context-realistic training simulators can significantly improve various aspects of operator training, especially safety and teamwork. The research provided an insight into the future of construction training simulator by indicating the importance and relevance of (1) collecting appropriate data, and (2) developing robust data-to-agent and data-to-physics methods.
•A framework for the context-realistic construction equipment training simulator is presented.•The framework integrates site layout data, mobility data, data-driven physics, and agent-based simulation.•Context-realistic simulators are shown to be promising for improving safety and teamwork of operator training.
Sustainability is becoming a key factor in the decision-making process of infrastructure projects throughout their lifecycles. In particular, the Environmental Impact Assessment (EIA) in the design ...phase is becoming a matter of significant importance, for both public and private sectors, given the long-term impacts of design decisions on the environmental performance of infrastructure projects. Traditionally, EIA is performed by a sustainability expert at the end of the design cycle, by which time the modification of design is both costly and time-consuming. In recent years, Building Information Modelling (BIM) is leveraged to better integrate EIA with the design practices. However, there are several limitations with how this integration is approached: (1) EIA is normally performed by software other than the one used for the design. This renders the continuous EIA based on incomplete BIM models difficult; (2) there is a lack of explicit data structure for the integration of EIA and BIM data. This limits the interoperability and flexibility of the EIA tools in terms of accommodating to different EIA databases; (3) in the majority of the cases the integration of EIA and BIM is not bidirectional, which results in the incapacity of the designers to immediately visualize the results of EIA in the design platform and to track the progress of the design in terms of EIA; and (4) the BIM-based EIA has rarely been implemented in an infrastructure project. Therefore, this research aims to develop a continuous BIM-based EIA for infrastructure projects that utilizes an explicit data structure to (1) systematically integrate data from various sources, and (2) enable bidirectional data exchange between BIM and EIA. The framework allows designers to run an automated EIA at any point in the design stage and immediately assess the Environmental Impact Score (EIS) of their design choices. A prototype is developed and tested on a case study to indicate the feasibility of the proposed framework. The framework is assessed in terms of functionality, ease of use, scalability, and contribution to raising sustainability consciousness through a workshop with experts. It is shown that the framework is able to quickly provide designers with accurate information about the potential environmental impact of all objects in infrastructure design projects. The workshop with experts showed that the tool clearly makes it easier to perform EIA compared to the existing, highly fragmented, process. This allows the design team to use this assessment on the same level as other design parameters in the decision-making process.
•A framework for BIM-based automated environmental impact assessment of infrastructure projects is developed.•A data structure for BIM-based environmental impact assessment is proposed.•The changes in the design workflow are analyzed.•A prototype is developed and tested for a real infrastructure project.•It is demonstrated that the framework raises the sustainability consciousness of the design team.
Although learning from projects has gained much importance in research and practice, progress in understanding and improving inter-project learning appears to be slight. We argue that the adoption of ...a sender/receiver approach limits the learning effectiveness in project-based organisations. Drawing upon the notion of learning as a social activity embedded in an organisational context, we develop the argument that learning from projects takes place within projects rooted in the historical, organisational and cultural context of previous and current projects. We underpin our argument with results from a multiple-case study on learning in construction organisations. We show that learning cannot be segregated from immediate practice and occurs when individuals engage in project work. Particularly the orientation towards project goals and project-overarching ambitions or trajectories can serve as contextual binder for learning in and between projects.
•The transfer of knowledge from one project to another project via several channels is impeded by characteristics that seem inherent to the contextual nature of project-based organisations.•An approach of learning between projects should consider the individual, social and organisational context through which projects are formed and which is constantly produced by project activities.•The goal orientation of project-based activities can and should serve as a contextual binder between projects, giving the social interaction within projects focus and orientation for the learning from projects.•If projects are perceived as sender/receiver islands, then lessons learned remain “messages in bottles” – freely afloat on the ocean of knowledge, arriving at new shores by chance.
Purpose
The purpose of this paper is to show that for frequently procuring public clients: the reasoning behind the use of procurement instruments is a process in its own right that requires ...managerial and scientific attention; modeling this process contributes to making sensible procurement choices; and managing this process is a relevant factor in the client’s development toward strategic procurement.
Design/methodology/approach
A model is developed to conceptualize the reasoning behind procurement instruments. Using this model in a case study, the reasoning behind the evolution of a particular procurement instrument as applied by a public infrastructure management organization is reconstructed.
Findings
The case study results show that an initially explicitly formulated set of main reasons for operating a qualification system can implicitly evolve over time into a different set of reasons. From a managerial point of view, explication of implicit reasons is important to both avoid the risk that the real value of the procurement instrument remains undetected as well as properly assess its strategic alignment with higher level strategies. The conceptual model proves to be a useful tool to support that.
Originality/value
Bringing the reasoning behind the use of procurement instruments to the fore, this study explores an area of construction procurement research that is positioned between the disciplines of purchasing and supply management, knowledge management and strategic management.
Bridge infrastructure managers are facing multiple challenges to improve the availability and serviceability of ageing infrastructure, while the maintenance planning is constrained by budget ...restrictions. Many research efforts are ongoing, for the last few decades, ranging from development of bridge management system, decision support tools, optimisation models, life cycle cost analysis, etc. Since transport infrastructures are deeply embedded in society, they are not only subject to technical requirements, but are required to meet the requirements of societal and economic developments. Therefore, bridge maintenance planning should accommodate multiple performance goals which need to be quantified by various performance indicators. In this paper, an application of Multi-Attribute Utility Theory (MAUT) for bridge maintenance planning is illustrated with a case study of bridges from the Netherlands road network. MAUT seeks to optimise multiple objectives by suggesting a trade-off among them and finally assigns a ranking to the considered bridges. Moreover, utility functions of MAUT appropriately account for the involved uncertainty and risk attitude of infrastructure managers. The main contribution of this study is in presenting a proof-of-concept on how MAUT provides a systematic approach to improve the decision-making of maintenance planning by making use of available data, accommodating multiple performance goals, their uncertainty, and preferences of infrastructure managers.