Mahshid Hatamzad disputerer for ph.d.-graden i ingeniørvitenskap / Mahshid Hatamzad will defend her thesis for the PhD degree in Engineering Science.
Mahshid Hatamzad disputerer for ph.d.-graden i ingeniørvitenskap og vil offentlig forsvare avhandlingen / Mahshid Hatamzad will defend her thesis for the PhD degree in Engineering Science:
“Data-driven Approach for Winter Road Safety and Maintenance”.
Avhandlingen er tilgjengelig her (lenke kommer) / The doctoral thesis is available here (link will be available soon).
Auditoriet er åpent for publikum. Disputasen vil også bli strømmet. Opptak av disputasen vil være tilgjengelig i en måned. / The auditorium is open to the public. The defense will be streamed. A recording of the defense will be available for one month.
Prøveforelesningen starter kl. 10:15 / The trial lecture starts at 10:15. Tittel kommer/ title will be available soon:
“”.
Disputasen starter kl. 12:15 / The defense starts at 12:15.
Prøveforelesning strømmes her, disputas strømmes her / The trial lecture will be streamed here, and defense will be streamed here.
Sammendrag av avhandlingen / Summary of the thesis:
A novel development in information and communication technology has formed the era of digitalization in various industries, including transportation. In recent years, significant and continuous improvements in sensors have brought the concept of intelligent transportation to the table. However, deploying intelligent transportation (e.g., marine, railway, and road) is challenging, especially in Nordic countries, where the winter is long, and the weather could be extremely cold. One of the main challenges in winter road transportation is enhancing traffic safety and clearing the roads of ice and snow (anti-icing, deicing, and plowing operations), which is called winter road maintenance (WRM). A lack of accurate information to perform good quality WRM operations can lead to reduced safety, extra expenses, and negative environmental impacts. Therefore, obtaining data-driven insights from the past to predict future patterns about road and weather information helps decision-makers to plan and perform efficient WRM. Hence, this thesis uses historical data (from different sources), analysis, and machine learning algorithms to:
• Analyzing and modeling severe road incidents: Historical weather-related data from road weather stations was used for analysis, classification, and modeling of severe road accidents to attain future insights for decision-making to improve winter road safety and WRM.
• Identifying the importance of weather-related variables affecting winter incident severity classification: analyzing historical data and feature importance approach helped to identify the importance of variables affecting the classification of severe winter road accidents in clodclimate areas, which can build a foundation to improve classification and prediction models for improving winter road safety.
• Finding a correlation between severe winter road incidents and road and weather conditions: utilizing data from multiple sources to explore the relationship between winter road accidents with personal injuries, road surface temperature, road surface conditions, and winter precipitation (snow, freezing rain, and sleet).
• Predicting road surface temperature: analyzing historical data from multiple sources for accurate prediction is essential for effective WRM operations such as salting to enhance road traffic safety in winter.
• To assess the uncertainty hidden in the prediction model for road surface temperature, a semiquantitative approach was utilized to evaluate the reliability of the proposed model and assess the uncertainty hidden in the background knowledge for tuning the model’s parameters.
• Predicting efficiency of WRM operations (salting): using a data envelopment analysis method to calculate the WRM efficiency (salting) using historical data from different sensors and road weather stations, and then constructing a model to classify and predict WRM efficiency classes and find the critical factors for WRM efficiency classification to derive future insights.
The findings of this Ph.D. thesis can improve winter road traffic safety and WRM efficiency operations in cold-climate areas by applying data analysis and machine learning algorithms. The models and insights proposed in this thesis can have useful applications in deriving data-driven future insights for safe winter road transportation and predictive WRM.
Veiledere / Supervisors:
Hovedveileder / Main supervisor:
As. Professor Geanette Cleotilde Polanco Pinerez, Department of Industrial Engineering, Faculty of Engineering Science and Technology, UiT The Arctic University of Norway
Biveileder / Co-supervisor:
Bedømmelseskomité / Evaluation committee:
Prøveforelesning og disputas ledes av prodekan for forskning, Bjarte Hoff / The trial lecture and defense are led by Vice Dean of research, Bjarte Hoff.
De som ønsker å opponere ex auditorio kan sende e-post til Bjarte Hoff / Opponents ex auditorio should contact Bjarte Hoff.