Edward Philip Smart Portrait

Dr Edward Smart

Principal Research Fellow

Edward.Smart@port.ac.uk

PhD Supervisor

Faculty of Technology School of Electrical and Mechanical Engineering Operational research and logistics Engineering - Electrical Electronic engineering Energy and Electronic Engineering Food Cultures in Transition research cluster Centre for Operational Research and Logistics

Biography

I am a Reader with over 15 years’ experience in the fields of data analysis and machine learning as well as a strong track record in industry-academia collaboration.

I am an effective collaborator, having won, or been a part of externally funded projects with staff from all 5 of the University's faculties.

I have been PI/Co-I on more than 20 Innovate UK, STFC and DASA bids worth nearly £6.8 million in total (worth over £3.4 million for the University). These grants have been in collaboration with academic partners (Southampton, Nottingham, UCLAN and Brunel) and industrial partners within the maritime, railway, aerospace, high performance computing and manufacturing sectors.

I was entered into REF 2014 (UoA 19) as an Early Career Researcher and I also undertook a large part of the research that formed one of the impact case studies. This case study was featured on the Universities Alliance . I was also entered into REF 2021 (UoA12) and submitted 2 out of the 4 impact case studies for this UoA.

Researcher development and support is a key interest. I sit on two working groups supporting the implementation of the Researcher Concordat within the University, as well as having previously served as Convenor of the Researchers' Network.

I received my PhD from 1024ºË¹¤³§ in 2011 and the MMath degree from the University of Reading in 2005. I previously worked at Clearswift Ltd where I used machine learning techniques to automatically identify certain types of images in email attachments. 

Research interests

My research interests focus on learning algorithms to detect rare or abnormal events from industrial data in situations where faulty data is sparse or non-existent. I am keen to apply such methods across several industry sectors as there are clear benefits to industry by improving maintenance methods through modern data analysis. Currently these methods are being applied in the manufacturing, computing and transport sectors of industry.