Development of an Explainable Traffic Flow Prediction Model Using Shapley Additive Explanation and Elliptic Bivariate Relationship

Geongyu Min; Dongju Ka; Sungyong Chung; Chungwon Lee

doi:10.7470/jkst.2023.41.2.239

All Issue

2023 Vol.41, Issue 2 Preview Page Next Page

Article

Development of an Explainable Traffic Flow Prediction Model Using Shapley Additive Explanation and Elliptic Bivariate Relationship SHAP과 Elliptic Bivariate Relationship을 이용한 설명가능한 교통량 예측모형 개발

30 April 2023. pp. 239-253

PDF XML

Abstract

Traffic flow prediction is essential for proactive and effective traffic management. The objective of this study is to apply the elliptic bivariate relationship, one of the traffic flow monitoring methods, to the traffic flow prediction model. Two parameters of the elliptic bivariate relationship, ϕ and λ, are employed in various machine learning-based prediction models. Subsequently, the feature importance of the parameters is assessed using SHAP (SHapley Additive exPlanations). The results based on empirical data indicate that the introduction of the elliptic parameters could reduce the prediction error in all models with prediction horizons of 5, 15, and 30 minutes. ϕ is found to have the highest feature importance among the explanatory variables in the models and λ showed similar or higher importance than the rest of the explanatory variables. These findings suggest that the elliptic parameters can be applied to various machine learning-based traffic flow prediction models.

Keywords

elliptic bivariate relationship

SHAP (SHapley Additive exPlanations)

traffic flow prediction model

machine learning

deep learning

교통관리전략 운영에 있어 교통류 소통상태를 파악하는 것은 매우 중요하며, 교통량 예측은 선제적이고 효과적인 교통관리를 위해 중요한 요소이다. 본 연구의 목적은 교통류 상태를 파악하는 새로운 방법 중 하나인 vehicle accumulation과 travel time 간의 elliptic bivariate relationship을 교통량 예측모형 개발에 적용하는 것이다. Elliptic bivariate relationship의 두 가지 파라미터인 ϕ와 λ를 다양한 유형의 머신러닝 및 딥러닝 기반 교통량 예측모형에 적용하고, SHAP(SHapley Additive exPlanations)을 이용하여 두 파라미터가 가지는 변수 중요도(feature importance)를 산정하였다. 본 연구는 개발된 각 예측모형을 고속도로 검지기 데이터에 적용하여 사례분석을 수행하였다. 5분 후, 15분 후, 30분 후 교통량 예측결과, 모든 예측모형에서 elliptic bivariate relationship 파라미터 적용을 통해 예측오차를 감소시킬 수 있는 것으로 나타났다. 또한, 변수 중요도를 산정한 결과, 예측모형에 적용된 설명변수 중 ϕ의 중요도가 가장 높은 것으로 나타났으며 λ의 중요도 역시 일반적인 교통류 변수와 유사하거나 다소 높은 수준인 것으로 확인되었다. 위 분석결과는 elliptic bivariate relationship 파라미터들이 다양한 머신러닝 기반의 교통량 예측모형에 적용될 수 있음을 보여준다.

키워드

elliptic bivariate relationship

SHAP(SHapley Additive exPlanations)

교통량 예측

머신러닝

딥러닝

References

Biau G., Scornet E. (2016), A Random Forest Guided Tour, Test, 25(2), 197-227. 10.1007/s11749-016-0481-7

Chen T., Guestrin C. (2016), Xgboost: A Scalable Tree Boosting System, In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794. 10.1145/2939672.2939785

Cheng T., Haworth J., Wang J. (2012), Spatio-temporal Autocorrelation of Road Network Data, Journal of Geographical Systems, 14(4), 389-413. 10.1007/s10109-011-0149-5

Chiabaut N., Faitout R. (2021), Traffic Congestion and Travel Time Prediction Based on Historical Congestion Maps and Identification of Consensual Days, Transportation Research Part C: Emerging Technologies, 124, 102920. 10.1016/j.trc.2020.102920

Davis G. A., Nihan N. L., Hamed M. M., Jacobson L. N. (1990), Adaptive Forecasting of Freeway Traffic Congestion, Transportation Research Record (1287).

Dunne S., Ghosh B. (2012), Regime-based Short-term Multivariate Traffic Condition Forecasting Algorithm, Journal of Transportation Engineering, 138(4), 455. 10.1061/(ASCE)TE.1943-5436.0000337

Ghosh B., Basu B., O’Mahony M. (2007), Bayesian Time-series Model for Short-term Traffic Flow Forecasting, Journal of Transportation Engineering, 133(3), 180-189. 10.1061/(ASCE)0733-947X(2007)133:3(180)

Hamed M. M., Al-Masaeid H. R., Said Z. M. B. (1995), Short-term Prediction of Traffic Volume in Urban Arterials, Journal of Transportation Engineering, 121(3), 249-254. 10.1061/(ASCE)0733-947X(1995)121:3(249)

Ka D., Chung K., Lee C. (2020), Real-Time Prediction of Vehicle Accumulation and Travel Times Over Extended Freeway Segments Using a Reproducible Elliptical Bivariate Relationship, Presented at the Transportation Research Board 99th Annual Meeting, Washington, DC.

Lundberg S. M., Lee S. I. (2017), A Unified Approach to Interpreting Model Predictions, Advances in Neural Information Processing Systems, 30.

Medsker L., Jain L. C. (Eds.) (1999), Recurrent Neural Networks: Design and Applications, CRC press. 10.1201/978142004917610566460PMC2232703

Oh Y., Chung K., Kang S., Jang K. (2019), Evaluating the Reproducible Relationship Between Travel Times and Vehicle Accumulation Over Extended Freeway Segments: Detecting Onset of Traffic Conditions Leading to a Marked Increase in Travel Times, Presented at the Transportation Research Board 98th Annual Meeting, Washington, DC.

Okutani I., Stephanedes YJ. (1984), Dynamic Prediction of Traffic Volume through Kalman Filtering Theory, Transportation Research, Part B, 18(1), 1-11. 10.1016/0191-2615(84)90002-X

Polson N. G., Sokolov V. O. (2017), Deep Learning for Short-term Traffic Flow Prediction, Transportation Research Part C: Emerging Technologies, 79, 1-17. 10.1016/j.trc.2017.02.024

Smith B. L., Demetsky M. J. (1997), Traffic Flow Forecasting: Comparison of Modeling Approaches, Journal of Transportation Engineering, 123(4), 261-266. 10.1061/(ASCE)0733-947X(1997)123:4(261)

Smola A. J., Schölkopf B. (2004), A Tutorial on Support Vector Regression, Statistics and Computing, 14(3), 199-222. 10.1023/B:STCO.0000035301.49549.88

Stathopoulos A., Karlaftis M. G. (2003), A Multivariate State Space Approach for Urban Traffic Flow Modeling and Prediction, Transportation Research Part C: Emerging Technologies, 11(2), 121-135. 10.1016/S0968-090X(03)00004-4

Vlahogianni E. I., Karlaftis M. G., Golias J. C. (2014), Short-term Traffic Forecasting: Where We are and Where We’re Going, Transportation Research Part C: Emerging Technologies, 43, 3-19. 10.1016/j.trc.2014.01.005

Williams B. M., Durvasula P. K., Brown D. E. (1998), Urban Freeway Traffic Flow Prediction: Application of Seasonal Autoregressive Integrated Moving Average and Exponential Smoothing Models, Transportation Research Record, 1644(1), 132-141. 10.3141/1644-14

Wu Y., Tan H., Qin L., Ran B., Jiang Z. (2018), A Hybrid Deep Learning Based Traffic Flow Prediction Method and Its Understanding, Transportation Research Part C: Emerging Technologies, 90, 166-180. 10.1016/j.trc.2018.03.001

Xu F., Lin Y., Huang J., Wu D., Shi H., Song J., Li Y. (2016), Big Data Driven Mobile Traffic Understanding and Forecasting: A Time Series Approach, IEEE Transactions on Services Computing, 9(5), 796-805. 10.1109/TSC.2016.2599878

Zhang D., Kabuka M. R. (2018), Combining Weather Condition Data to Predict Traffic Flow: A GRU-based Deep Learning Approach, IET Intelligent Transport Systems, 12(7), 578-585. 10.1049/iet-its.2017.0313

Information

Publisher :Korean Society of Transportation
Publisher(Ko) :대한교통학회
Journal Title :Journal of Korean Society of Transportation
Journal Title(Ko) :대한교통학회지
Volume : 41
No :2
Pages :239-253
Received Date : 2022-11-30
Revised Date : 2022-12-12
Accepted Date : 2022-12-29
DOI :https://doi.org/10.7470/jkst.2023.41.2.239

Journal of Korean Society of Transportation ISSN:1229-1366(Print) 2234-4217(Online) 대한교통학회지

All Issue