This presentation is a case study taken from the travel and holiday industry. Paxport/Multicom, based in UK and Sweden, have recently adopted a recommendation system for holiday accommodation bookings. Machine learning techniques such as Collaborative Filtering have been applied using Python (3.5.1), with Jupyter (4.0.6) as the main framework. Data scale and sparsity present significant challenges in the case study, and so the effectiveness of various techniques are described as well as the performance of Python-based libraries such as Python Data Analysis Library (Pandas), and Scikit-learn (built on NumPy, SciPy and matplotlib). The presentation is suitable for all levels of programmers.

Machine learning with Python

The road pavement conditions affect safety and comfort, traffic and travel times, vehicles operating cost, and emission levels. In order to optimize the road pavement management and guarantee satisfactory mobility conditions for all road users, the Pavement Management System (PMS) is an effective tool for the road manager. An effective PMS requires the availability of pavement distress data, the possibility of data maintenance and updating, in order to evaluate the best maintenance program. In the last decade, many researches have been focused on pavement distress detection, using a huge variety of technological solutions for both data collection and information extraction and qualification. This paper presents a literature review of data collection systems and processing approach aimed at the pavement condition evaluation. Both commercial solutions and research approaches have been included. The main goal is to draw a framework of the actual existing solutions, considering them from a different point of view in order to identify the most suitable for further research and technical improvement, while also considering the automated and semi-automated emerging technologies. An important attempt is to evaluate the aptness of the data collection and extraction to the type of distress, considering the distress detection, classification, and quantification phases of the procedure.

Pavement Distress Detection Methods: A Review

Cross-scene pavement distress detection by a novel transfer learning framework

Pavement asset management systems and technologies: A review

Adoption of machine learning technology for failure prediction in industrial maintenance: A systematic review

In recent years, there has been an increasing interest in the application of machine learning for the prediction of pavement performance. Prediction models are used to predict the future pavement c...

Machine learning approach for pavement performance prediction

The selection and use of technical parameters and performance indicators plays an essential role in the pavement management process. It is known that if more parameters are used, a more accurate ev...

Comprehensive performance indicators for road pavement condition assessment

Accurate pavement performance prediction models are essential to ensure optimal allocation of resources in maintenance management. These models are developed using inventory and monitoring data regarding pavement structure, climate, traffic, and condition. However, numerous road agencies have limited pavement data. Due to the inexistence of historical data, data collection frequency, and/or quality issues, the amount of data available for the development of performance models is reduced. As a result, the resource allocati on process is significantly undermined. This paper proposes a transfer learning approach to develop pavement performance prediction models in limited data contexts. The proposed transfer learning approach is based on a boosting algorithm. In particular, a modified version of the popular TrAdaBoost learning algorithm was used. To test the proposed transfer learning approach, a case study was developed using data from the Long-Term Pavement Performance (LTPP) database and from the Portuguese road administration database. The results of this work show that it is possible to develop accurate performance prediction models in limited data contexts when a transfer learning approach is applied. All the models resulting from this approach outperformed baseline models, especially in what regards long-term forecasts. The results also showed that the transfer learning models perform consistently over different time frames, with minor performance losses from one-step to multi-step forecasts. The findings of this study should be of interest to road agencies facing limited data contexts and aiming to develop accurate prediction models that can improve their pavement management practice.

Transfer learning for pavement performance prediction

During the last decades, the advent of Artificial Intelligence (AI) has been taking place in several technical and scientific areas. Despite its success, AI applications to solve real-life problems in pavement engineering are far from reaching its potential. In this paper, a Python machine learning library, scikit-learn, is used to predict asphalt pavement friction. Using data from the Long-Term Pavement Performance (LTPP) database, 113 different sections of asphalt concrete pavement, spread all over the United States, were selected. Two machine learning models were built from these data to predict friction, one based on linear regression and the other on regularized regression with lasso. Both models showed to be feasible and perform similarly. According to the results, initial friction plays an essential role in the way friction evolves over time. The results of this study also showed that scikit-learn can be a versatile tool to solve pavement engineering problems. By applying machine learning methods to predict asphalt pavements friction, this paper emphasizes how theory and practice can be effectively coupled to solve real-life problems in contemporary transportation.

Machine Learning for Pavement Friction Prediction Using Scikit-Learn

This paper proposes the application of the MACBETH approach for the development of a multi criteria value model to enable the prioritization of maintenance and rehabilitation interventions in the pavements of road networks according to technical, economic, and social aspects. A case-study design considering the Portuguese network was undertaken to illustrate the proposed approach. Using an interactive structuring process developed with the decision-maker, a set of five evaluation criteria related to maintenance costs, traffic, pavement condition, safety, and social equity were defined. The MACBETH approach was then used to build value functions and establish relative weights for the criteria. An application of the resulting model was made to a portfolio of possible intervention projects, prioritizing them according to the principle of benefit/cost maximization. The case-study demonstrated the application of the MACBETH approach and its suitability to solve the decision-making problem, thus proving the usefulness of this approach for road agencies with pavement management responsibilities.

/pdf/development-of-a-multi-criteria-decision-analysis-model-for-1x1a6ocufg.pdf

Pedro Marcelino

Papers

Machine learning approach for pavement performance prediction

Comprehensive performance indicators for road pavement condition assessment

Transfer learning for pavement performance prediction

Machine Learning for Pavement Friction Prediction Using Scikit-Learn

Development of a Multi Criteria Decision Analysis Model for Pavement Maintenance at the Network Level: Application of the MACBETH Approach