Search

Artificial intelligence (AI) has been widely used in robotics, automation, finance, healthcare, etc. However, using AI for decision-making in sustainable product lifecycle operations is still challenging. One major challenge relates to the scarcity and uncertainties of data across the product lifecycle. This paper aims to develop a method that can adopt the most suitable AI techniques to support decision-making for sustainable operations based on the available lifecycle data. It identifies the key lifecycle stages in which AI, especially reinforcement learning (RL), can support decision-making. Then, a generalised procedure of using RL for decision support is proposed based on available lifecycle data, such as operation and maintenance data. The method has been validated in a case study of an international vehicle manufacturer, combined with modelling and simulation. The case study demonstrates the effectiveness of the method and identifies that RL is the current most appropriate method for maintenance scheduling based on limited available lifecycle data. This paper contributes to knowledge by demonstrating an empirically grounded industrial case using RL to optimise decision-making for improved product lifecycle sustainability by effectively prolonging the product lifetime and reducing environmental impact., Funding Agencies|VinnovaVinnova [2017-01649]

Waste management in the Architecture, Engineering and Construction (AEC) industry has been a major research topic owing to the AEC industry being one of the top contributors of waste generation. However, research has primarily focused on new build and has neglected refurbishment projects which become relevant due to an aging building stock in Sweden and Europe. Various actors are involved in refurbishment projects which makes it important to study each actor as well as tensions between them. By using a service ecosystem perspective and relying on 38 interviews, this paper aims to investigate tensions and barriers between actors within the service ecosystem of CDW for refurbishment projects in Sweden. Based on the results, spatial barriers are the most mentioned barrier which also create the highest number of tensions between project and contractor mother firm, subcontractors, waste recycling firms and society/citizens. The majority of the tensions are found between projects and the contractor mother firms, followed by tensions between projects and clients and projects and subcontractors. Unlike previous studies investigating only one of the actors waste management practices, this paper contributes by investigating the interactions between seven different actors which is important for improving waste management practices in refurbishment projects., Funding Agencies|Mistra Closing the Loop, the Foundation for Strategic Environmental Research [160026]

Artificial intelligence (AI) has been widely used in robotics, automation, finance, healthcare, etc. However, using AI for decision-making in sustainable product lifecycle operations is still challenging. One major challenge relates to the scarcity and uncertainties of data across the product lifecycle. This paper aims to develop a method that can adopt the most suitable AI techniques to support decision-making for sustainable operations based on the available lifecycle data. It identifies the key lifecycle stages in which AI, especially reinforcement learning (RL), can support decision-making. Then, a generalised procedure of using RL for decision support is proposed based on available lifecycle data, such as operation and maintenance data. The method has been validated in a case study of an international vehicle manufacturer, combined with modelling and simulation. The case study demonstrates the effectiveness of the method and identifies that RL is the current most appropriate method for maintenance scheduling based on limited available lifecycle data. This paper contributes to knowledge by demonstrating an empirically grounded industrial case using RL to optimise decision-making for improved product lifecycle sustainability by effectively prolonging the product lifetime and reducing environmental impact., Funding Agencies|VinnovaVinnova [2017-01649]

Waste management in the Architecture, Engineering and Construction (AEC) industry has been a major research topic owing to the AEC industry being one of the top contributors of waste generation. However, research has primarily focused on new build and has neglected refurbishment projects which become relevant due to an aging building stock in Sweden and Europe. Various actors are involved in refurbishment projects which makes it important to study each actor as well as tensions between them. By using a service ecosystem perspective and relying on 38 interviews, this paper aims to investigate tensions and barriers between actors within the service ecosystem of CDW for refurbishment projects in Sweden. Based on the results, spatial barriers are the most mentioned barrier which also create the highest number of tensions between project and contractor mother firm, subcontractors, waste recycling firms and society/citizens. The majority of the tensions are found between projects and the contractor mother firms, followed by tensions between projects and clients and projects and subcontractors. Unlike previous studies investigating only one of the actors waste management practices, this paper contributes by investigating the interactions between seven different actors which is important for improving waste management practices in refurbishment projects., Funding Agencies|Mistra Closing the Loop, the Foundation for Strategic Environmental Research [160026]

The exact distribution of a classification function is often complicated to allow for easy numerical calculations of misclassification errors. The use of expansions is one way of dealing with this difficulty. In this paper, approximate probabilities of misclassification of the maximum likelihood-based discriminant function are established via an Edgeworth-type expansion based on the standard normal distribution for discriminating between two multivariate normal populations.

The exact distribution of a classification function is often complicated to allow for easy numerical calculations of misclassification errors. The use of expansions is one way of dealing with this difficulty. In this paper, approximate probabilities of misclassification of the maximum likelihood-based discriminant function are established via an Edgeworth-type expansion based on the standard normal distribution for discriminating between two multivariate normal populations.

Time-dependent Origin-Destination (OD) matrices are an essential input to transportation models. A cost-efficient and widely used approach for estimating OD matrices involves the exploitation of flow counts from stationary traffic detectors. This estimation approach is also referred to as assignment-based OD matrix estimation because, typically, Dynamic Traffic Assignment (DTA) models are used to map the OD matrix to the link flows. The conventional DTA establish a complex non-linear relationship between the demand, and the link flows, adding an inherent complexity to the OD matrix estimation problem. In this paper, attempting to exploit the growing availability of Floating-Car Data (FCD), we suggest a solution approach that is based on a Data-Driven Network Assignment (DDNA) mechanism. The DDNA utilises the FCD from probe vehicles to capture congestion effects, providing a linear mapping of the OD matrix to the link flow observations. We present the results of two synthetic-data experiments that serve as proof of concept, indicating that if FCD are available, the computationally costly DTA may not be necessary for solving the OD matrix estimation problem., Funding Agencies|Swedish Transport Administration [TRV2018/132473, TRV2021/22404]; Swedish Energy Agency [46963-1]

Time-dependent Origin-Destination (OD) matrices are an essential input to transportation models. A cost-efficient and widely used approach for estimating OD matrices involves the exploitation of flow counts from stationary traffic detectors. This estimation approach is also referred to as assignment-based OD matrix estimation because, typically, Dynamic Traffic Assignment (DTA) models are used to map the OD matrix to the link flows. The conventional DTA establish a complex non-linear relationship between the demand, and the link flows, adding an inherent complexity to the OD matrix estimation problem. In this paper, attempting to exploit the growing availability of Floating-Car Data (FCD), we suggest a solution approach that is based on a Data-Driven Network Assignment (DDNA) mechanism. The DDNA utilises the FCD from probe vehicles to capture congestion effects, providing a linear mapping of the OD matrix to the link flow observations. We present the results of two synthetic-data experiments that serve as proof of concept, indicating that if FCD are available, the computationally costly DTA may not be necessary for solving the OD matrix estimation problem., Funding Agencies|Swedish Transport Administration [TRV2018/132473, TRV2021/22404]; Swedish Energy Agency [46963-1]

Time-dependent Origin-Destination (OD) matrices are an essential input to transportation models. A cost-efficient and widely used approach for estimating OD matrices involves the exploitation of flow counts from stationary traffic detectors. This estimation approach is also referred to as assignment-based OD matrix estimation because, typically, Dynamic Traffic Assignment (DTA) models are used to map the OD matrix to the link flows. The conventional DTA establish a complex non-linear relationship between the demand, and the link flows, adding an inherent complexity to the OD matrix estimation problem. In this paper, attempting to exploit the growing availability of Floating-Car Data (FCD), we suggest a solution approach that is based on a Data-Driven Network Assignment (DDNA) mechanism. The DDNA utilises the FCD from probe vehicles to capture congestion effects, providing a linear mapping of the OD matrix to the link flow observations. We present the results of two synthetic-data experiments that serve as proof of concept, indicating that if FCD are available, the computationally costly DTA may not be necessary for solving the OD matrix estimation problem., Funding Agencies|Swedish Transport Administration [TRV2018/132473, TRV2021/22404]; Swedish Energy Agency [46963-1]

The exact distribution of a classification function is often complicated to allow for easy numerical calculations of misclassification errors. The use of expansions is one way of dealing with this difficulty. In this paper, approximate probabilities of misclassification of the maximum likelihood-based discriminant function are established via an Edgeworth-type expansion based on the standard normal distribution for discriminating between two multivariate normal populations.