|
Prof. Andrew Evans Prof.AndrewEvans
Lloyd's Register Professor of Transport Risk Management
|
|
|
Contact Information
|
E-mail:
|
|
a.evans@imperial.ac.uk
|
|
Address:
|
|
Room 435, Skempton Bldg. Imperial College London SW7 2AZ |
|
Telephone:
|
|
+44(0)20 7594-6043 |
|
Fax:
|
|
+44(0)20 7594-6102 |
|
Related Webpage:
|
|
|
Academic Information
| Teaching Activities |
|
Research Interests |
| Please check the MSc
website. |
|
|
CV:
Download file
Home, Top of page and photo, Publications, Projects, news items
Summary of Publications 
This is not necessarily a complete list of publications by the staff member.Publications before 1999 are not included.Please download the CV for full details.
Publication Format: Working Paper
Publication Format: Journal - In press
Publication Format: Journal
Publication Format: Book
Bichou, K, Bell, M.G.H. and Evans, A, 2007, Risk Management in Port Operations, Logistics and Supply Chain Security, Informa, London, ISBN:978-1-84311-655-4
|
Publication Format: Book Chapters
Bichou, K and Evans, A, 2007, Maritime Security and Regulatory Risk-Based Models: Review and Critical Analysis, Edited by Bichou, K, Bell, M.G.H. and Evans, A, Risk Management in Port Operations, Logistics and Supply Chain Security, Informa, London
|
Publication Format: Conference Proceedings (Published)
Publication Format: Conference Proceedings (unpublished)
Publication Format: PhD Thesis
Home, Top of page and photo, Publications, Projects, news items
Summary of Projects
Fatal train accidents on Britain's main line railways
A.W. Evans
Andrew Evans continues to produce an annual analysis of fatal train accidents on the British main line railway network with papers at the end of 2003 and 2004 based on data from 1967. The work now covers collisions between trains and road vehicles as well as train collisions, derailments and overruns.
The effect of safety interventions and economic growth on motorcycle crashes and fatalities
Law, Teik-Hua (PhD project. Supervisors: Professor A W Evans and Dr R B Noland)
In developing countries, motorcycle usage is associated with income levels, and in general, a motorcycle is relatively cheaper and is often the only affordable means of transport for poor people. However, the growing use of motorcycles is associated with a higher death or injury risk compared to other forms of transportation. In south-east Asia, there are several countries with a large proportion of motorcycles whose growth in numbers has been associated with a large rise in fatal injuries. The objective of this research is to empirically investigate the effects of motorcycle safety interventions and economic growth on motorcycle crashes and fatalities. Specifically, this research seeks to examine associations between economic growth and motorcyclist fatalities and how this interacts with policies to reduce fatalities, such as helmet legislation and daytime running lights.
The Effects of Political governance, Policy Measures and Economic Growth on Motorcyclist and Total Road Deaths (PhD project)
Law Teik Hua (PhD project. Supervisors: Professor A W Evans and Dr R B Noland)
Recently, several analysts have estimated the relationship between per capita income and road deaths. All of these studies concluded that there is an inverted-U shape relationship between income and road deaths, which is similar to the one Kuznets in 1955 suggested existed between income inequality and per capita income (also called Environmental Kuznets Curve or EKC). However, an important criticism of previous studies is that although they describe the inverted U-shaped relationship between road deaths and economic growth they do not explain the underlying mechanism driving this relationship. A possible explanation of this relation is that often at relatively low levels of per capita income, an increase in income will lead to a rapidly growing number of motor vehicles in the population. This causes a rapid increase in the threat to non-motorized road users (pedestrians and bicyclists), causing total road deaths to rise. Nevertheless, starting from a certain level of income, concerns about road safety have become increasingly pronounced, which makes it possible to create new institutions and regulations to enhance road safety, result in the levelling off and gradual decline in road deaths.
Several issues arise from this line of research.
(1) Do road deaths indeed follow an “EKC”, first rising then falling as income increases?
(2) At what level of per capita income is the turning point?
(3) How does this relationship varies across different type of road user deaths?
(4) Is deterioration of road safety at lower income levels due to increases in a higher degree of mobility and a greater demand for transport services?
(5) Is road safety improvement at higher income levels automatic, or due to primarily to transport mode composition changes, motor vehicle-related technological and medical technology advances, or improvement in education?
(6) Does the degree and form of democracy, and the quality of governance affect the level of road safety quality enjoyed by a country independently of that country’s income?
(7) Are civil and political freedoms good or bad for the road safety?
(8) How can the process be accelerated so that developing and transition economies can attain the same improved road safety conditions enjoyed by developed market economies?
This research contributes to the literature by conducting three empirical investigations and the more specific objectives are as follows:
(i) To study the underlying mechanism of the inverted-U shape relationship between total road deaths and economic growth. The main components of income, scale and abatement effects are examined to determine their relative roles in the road deaths and economic growth relationship.
(ii) To examine the effect of political development (Political system: democracy or autocracy, democratic system, and civil liberties) and the quality of governance (corruption level, rule of law, and regulatory quality) on road deaths.
(iii) To identify potential road safety policy programs in developing countries that can minimize road death risk while achieving sustainable economic growth and socioeconomic development.
All research students are initially registered for an MPhil and transfer from MPhil to PhD when their progress justifies this. This process requires the student to present a transfer paper and have this accepted. Mr Law transferred to PhD in June 2007. Mr Law has also produced two conference papers based on his work.
Modelling the Severity of Fatal Train Accidents in Great Britain
Mohammed Quddus and Andrew Evans
This project performed a new and more advanced statistical analysis of the severity of train collisions, derailments and overruns on the national railway system of Great Britain using Andrew Evans’ dataset of the 80 fatal accidents in 1967 to 2005. The statistical model used was ‘ordered probit’ models. The aim of the analysis was to investigate what factors influence the severity of accidents. The measure of severity is the number of fatalities in accidents, and model investigates the influence the impact speed of the train, type of rolling stock, whether the train is multiple-unit or locomotive hauled, type of accident and date of the accident. For fitting the ordered probit model, the accidents were classified into three levels of severity: low fatality accidents (1 to 3 fatalities), medium fatality accidents (4 to 10 fatalities) and high fatality accidents (more than 10 fatalities).
The results suggest that the statistically significant factors influencing severity are the impact speed of the train, the type of rolling stock (Mark 1 or post-Mark 1), and the type of train (either multiple units or locomotive-hauled) in both models. Interestingly, the accidents due to signals passed at danger were not found to be statistically significantly more or less severe the other accidents in either model. The inclusion of the time trend variable (using both linear and exponential growth trends) in the models indicated that the severity of fatal train accidents did not change over time. The estimation of marginal effects indicates that each additional 10 km/h of impact speed decreases the probability of a low fatality accident by 0.08 and increases the probability of a medium fatality accident by 0.08 and the probability of a high fatality accident by 0.005, while all other variables in the model remain unchanged. The probability of a medium fatality accident is 0.28 lower for modern rolling stock than for older rolling stock and the probability of a medium fatality accident is 0.32 higher for multiple units train than for locomotive-hauled train.
A paper on this work was presented in poster form at the 86th Transportation Research Board Annual Conference in January 2007 in Washington DC, and submitted for publication in the Transportation Research Record.
Modelling the impacts of security on container-port operational efficiency and performance benchmarking
Khalid Bichou (PhD project. Supervisors: Professor A W Evans, Professor M G H Bell and visiting Professor R Cochrane)
The research project attempts to model the ex-post impacts of the new security measures on world’s container port operational efficiency and benchmarking. In other words, we seek to know the extent to which the procedures and requirements induced by the new security measures have impacted the efficiency of global container port operations both over time and vis-à-vis each other. This research seeks to adopt an approach that incorporates within a valid framework of analysis the association of security with operational efficiency, valid measures and techniques for benchmarking container-port efficiency, and appropriate tools for modelling security impacts and processes.
A panel data set for 65 container terminals and 45 container ports from 2000 to 2006 is used, giving 660 decision-making units (DMUs). Primary and secondary data are sourced from individual ports and terminals (both open and survey-based sources) and from subscribed databases of professional and trade journals. In order to equally account for container terminal operational configurations and the multi-input / multi-output nature of container port production, we apply a combination of operations engineering, process modelling and analytical benchmarking techniques, respectively cost engineering methods, the Integrated computer aided manufacturing DEFinition (IDEF0) for activity modelling, and Data Envelopment Analysis (DEA) for efficiency benchmarking. We then estimate a DEA-based Malmquist total factor productivity (TFP) index to measure and decompose productivity changes at both terminal and port levels. We use different sets of input and output variables so that we can account for the variations in terminal operating systems and the spatial scope of each security regulation. We also run several DEA models in order to capture the influence of exogenous variables on port and terminal efficiency.
Partial and preliminary results (compiled from 15 container terminals) provide evidence of a high correlation between efficiency levels and terminal operating configurations. They also substantiate the influence of exogenous factors (traffic and cargo mix) on operational efficiency as well as the lagging effect of incremental investments on productive efficiency. There are however mixed results regarding the impacts of security on operational efficiency. While some ports and terminals depict negative productivity change following the introduction of security measures, others especially large terminals show a growth in productive efficiency implying that security measures have been more detrimental to small and medium sized ports. However and beyond scale properties, much of the impact (negative or positive) is being explained by terminal operating processes and procedures (terminal working hours, gate closure procedure, free storage policy, etc.) as captured by our empirical model. Even though, one should be reminded that such results stem only from partial and preliminary analysis and may not provide a true or full quantification of the security impacts. Both efficiency scores and productivity change indexes are relative to the sample being analysed, and one must not jump into conclusions until the full results become available.
Home, Top of page and photo, Publications, Projects, news items
Summary of News Items
April 2007
Prof. Andrew Evans participated in a discussion on BBC Radio 4 on rail privatisation and safety. The show "More or Less" was broadcast on Monday 30 April, featuring a discussion between the presenter Andrew Dilnot, Christian Wolmar and Andrew Evans. More details can be found at http://news.bbc.co.uk/1/hi/programmes/more_or_less/6607991.stm
June 2005
The Centre for Transport Studies hosted an international two-day workshop (25 and 26 May) on aviation safety. The workshop brought together the world's civil aviation authorities and air navigation service providers to look at how records of previous incidents and accidents can be used to prevent accidents in the future. The workshop, with financial assistance from Lloyd's Register, was the first to bring aviation authorities together to discuss and learn from their collective experience of safety incidents and to define a research programme in collaboration with a major research institution. Members of the organising Committee included Dr Washington Ochieng, Dr Arnab Majumdar, Professor Andrew Evans, Jackie Sime, Miss Branka Subotic and Miss Marie-Dominique Dupuy. More information can be found at http://www.imperial.ac.uk/p6395.htm
October 2004
Prof. Andrew Evans represented UK as one of two UK speakers at UK/Korea Rail Research Seminar at Seoul in October 2004.
July 2004
Prof. Andrew Evans gave Inaugural Lecture and First Lloyd’s Register Lecture in Transport Risk Management at Imperial College, entitled “Rail safety and rail privatisation in Britain”.
May 2004
Professor Andrew Evans will give the First Lloyds Register Transport Risk Management Lecture and his Inaugural Lecture in the Clore Lecture Theatre at Imperial College London at 17:30 on Wednesday 16 June 2004. The title will be "Rail safety and rail privatisation in Britain". The Lecture is open to all. Anyone who wishes to come is asked to contact Pamela Michael at p.michael@imperial.ac.ukHome, Top of page and photo, Publications, Projects, news items
|
