Predicting Road Accidents Based on Current and Historical Spatio-temporal Traffic Flow Data - Computational Logistics - page 96

online and feedback about the accident risk is fed back to the controller in real time so
that preventive action can be taken. Therefore, finally, an interface will be created to
feedback the prediction of the outcome of current traffic flow data in real time to an
agency, such as the Welsh Traffic Technology Consultancy or to an automated system
to accordingly take anticipative actions, such as setting variable speed limits. An in-
teresting extension of this system would be to predict traffic flow based on actions
such as changing speed limits and then re-analysing the risk of an accident occurring.
This can be done again using CBR by comparing the changes in flow data due to
anticipative actions to similar changes in the past and their outcomes.
Acknowledgements. This work was funded by an EPSRC Knowledge Transfer Se-
condment programme allocated to the University of Nottingham. The work was done
in collaboration with Innovation Works, EADS . Valuable guidance and traffic data
was provided by the Welsh Traffic Technology Consultancy (WTTC), Wales.
References
1. https://www.gov.uk/government/statistical-data-sets/ras10-
reported-road-accidents
2. Elvik, R.: How much do road accidents cost the national economy? Accident Analysis &
Prevention 32, 849-851 (2000)
3. Kolodner, J.: Case-Based Reasoning. Morgan Kaufmann, San Francisco (1993)
4. Golob, T.F., Recker, W.W.: Relationships among urban freeway accidents, traffic flow,
weather, and lighting conditions. Journal of Transportation Engineering 129, 342-353
(2003)
5. Golob, T.F., Recker, W.W., Alvarez, V.M.: Freeway safety as a function of traffic flow.
Accident Analysis & Prevention 36, 933-946 (2004)
6. Taehyung, K., Hyoungsoo, K., Lovell, D.J.: Traffic flow forecasting: overcoming memo-
ryless property in nearest neighbor non-parametric regression. In: Proceedings of the Intel-
ligent Transportation Systems, pp. 965-969. IEEE (2005)
7. Lee, C., Abdel-Aty, M.A.: Two-level nested Logit model to identify traffic flow parame-
ters affecting crash occurrence on freeway ramps. Transportation Research Record 2083,
145-152 (2008)
8. Garber, N., Ehrhart, A.: Effect of speed, flow, and geometric characteristics on crash fre-
quency for two-lane highways. Transportation Research Record: Journal of the Transporta-
tion Research Board 1717, 76-83 (2000)
9. Cheol, O., Jun-Seok, O., Ritchie, S.G.: Real-time hazardous traffic condition warning sys-
tem: framework and evaluation. IEEE Transactions on Intelligent Transportation Sys-
tems 6, 265-272 (2005)
10. Saltz, J.H., Teodoro, G., Pan, T., Cooper, L.A.D., Kong, J., Klasky, S., Kurc, T.M.: Fea-
ture-based analysis of large-scale spatio-temporal sensor data on hybrid architectures. In-
ternational Journal of High Performance Computing Applications (2013)
11. George, B., Kim, S.: Spatio-temporal networks: An introduction. In: Spatio-temporal Net-
works, pp. 1-6. Springer, New York (2013)
12. Treiber, M., Kesting, A.: Validation of traffic flow models with respect to the spatiotem-
poral evolution of congested traffic patterns. Transportation Research Part C: Emerging
Technologies 21, 31-41 (2012)