Skip to main content
SpringerLink
Log in

Bayesian combined neural network for traffic volume short-term forecasting at adjacent intersections

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

Undoubtedly, one of the greatest issues nowadays is congestion. To face such problem, forecasting of traffic is required. Bayesian combined neural network (BCNN) is applied to four different locations in Kuwait (Cairo Street, Riyadh Street, Maghreb Road and Istiqlal Road) to predict the short-term traffic volume at the middle section due to traffic flow from adjacent intersections. All data were collected for a period of 1 week over 15-min observation intervals using loop detectors. In addition to time-series responses and regression plots, mean square error (MSE) has been used to validate the network performance after data normalization. In comparison with MSE and R values, both values were slightly less precise during weekdays compared to weekends. After standardizing, the average MSE during weekdays was 0.003468 and regression (R) was 0.98113 for the four streets. For weekends model, the average MSE was 0.003563 and regression (R) was 0.97374 for the four streets. Istiqlal Street weekday model was the best model that fits the information among all the four models; as it has the smallest MSE value equivalent to 0.0010087 and the highest R value of 0.9959. BCNN model has achieved outstanding prediction performance with great potential to be generalized for various locations at different times of the day. These results can allow transportation planners to forecast traffic congestions and take prior measures to avoid them. Further modeling can assist in studying factors causing intersection congestions.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19
Fig. 20
Fig. 21

Similar content being viewed by others

References

  1. Vlahogianni EI, Golias JC, Karlaftis MG (2004) Short-term traffic forecasting: overview of objectives and methods. Transp Rev 24(5):533–557

    Article  Google Scholar 

  2. Karlafris M, Vlahogianni E (2011) Statistical methods versus neural networks in transportation research: differences, similarities, and some insights. Transp Res Part C Emerg Technol 19(3):387–399

    Article  Google Scholar 

  3. Yang J, Ma J (2019) Feed-forward neural network training using sparse representation. Expert Syst Appl 116:255–264

    Article  Google Scholar 

  4. Lieu HC (2000) Traffic estimation and prediction system. TR News 208:3–6

    Google Scholar 

  5. Chen H, Grant-Muller S (2001) Use of sequential learning for short-term traffic flow forecasting. Transp Res Part C 9:319–336

    Article  Google Scholar 

  6. Head LK (1995) Event-based short-term traffic flow prediction model. Transp Res Board 1510:45–52

    Google Scholar 

  7. Kirby H, Dougherty M, Watson S (1997) Should we use neural networks or statistical models for short term motorway forecasting. Int J Forecast 13:45–50

    Article  Google Scholar 

  8. Ishak S, Al-Deek H (2002) Performance of short-term time series traffic prediction model. J Transp Eng 128(6):490–498

    Article  Google Scholar 

  9. Duan H, Xiao X, Xiao Q (2019) An inertia grey discrete model and its application in short-term traffic flow prediction and state determination. Neural Comput Appl. https://doi.org/10.1007/s00521-019-04364-w

    Article  Google Scholar 

  10. Hu W, Wang H, Yan L (2014) An actual urban traffic simulation model for predicting and avoiding traffic congestion. In: 17th International IEEE conference on intelligent transportation systems (ITSC), Qingdao, pp 2681–2686

  11. Ma C, He R (2019) Green wave traffic control system optimization based on adaptive genetic-artificial fish swarm algorithm. Neural Comput Appl 31:2073–2083

    Article  Google Scholar 

  12. Smith BL, Demetsky MJ (1997) Traffic flow forecasting: comparison of modelling approaches. J Transp Eng 123(4):261–266

    Article  Google Scholar 

  13. Vythoulkas PC (1993) Alternative approaches to short-term traffic forecasting for use in driver information systems. In: Transportation and traffic theory, proceedings of the 12th international symposium on traffic flow theory and transportation, Berkley, CA, 21–23 July, a symposium in honor of Gordon F. Newell

  14. Abdulhai B, Porwal H, Recker W (1999) Short-term freeway traffic flow prediction using genetically optimized time-delay-based neural networks. UCB, UCB-ITS-PWP–99–1. Institute of Transportation Studies, University of California, Berkeley

  15. Levin M, Tsao Y-D (1980) On forecasting freeway occupancies and volumes. Transp Res Rec 773:47–49

    Google Scholar 

  16. Lin W-H, Lu Q, Dahlgren J (2002) A dynamic procedure for short-term prediction of traffic congestion. In: Proceedings of the 81st transportation research record annual meeting, Washington, DC

  17. Dougherty MS, Cobbet MR (1997) Short-term inter-urban traffic forecasts using neural networks. Int J Forecast 13:21–31

    Article  Google Scholar 

  18. Song Z, Guo Y, Wu Y, Ma J (2019) Short-term traffic speed prediction under different data collection time intervals using a SARIMA-SDGM hybrid prediction model. PLoS ONE. https://doi.org/10.1371/journal.pone.0218626

    Article  Google Scholar 

  19. Florio L, Mussone L (1996) Neural network models for classification and forecasting of freeway traffic flow stability. Control Eng Pract 4(2):153–164

    Article  Google Scholar 

  20. Innamaa S (2000) Short-term prediction of traffic situation using MLP-neural networks. In: Proceedings of the 7th world Congress on intelligent transportation systems, Turin, Italy C, vol 5, No. 5, pp 287–300

  21. Davis GA, Niham NL, Hamed MM, Jacobson LN (1991) Adaptive forecasting of freeway traffic congestion. Transp Res Rec 1287:29–33

    Google Scholar 

  22. Rehborn H, Klenov SL, Koller M (2019) Traffic prediction of congested patterns. In: Kerner BS (ed) Complex dynamics of traffic management. Springer, Berlin, pp 501–557. https://doi.org/10.1007/978-1-4939-8763-4_564

    Chapter  Google Scholar 

  23. Zou W, Xia Y (2019) Back propagation bidirectional extreme learning machine for traffic flow time series prediction. Neural Comput Appl 31:7401–7414

    Article  Google Scholar 

  24. Lyons GD, McDonald M, Hounsell NB, Williams B, Cheese J, Radia B (1996) Urban traffic management: the viability of short-term congestion forecasting using artificial neural networks. In: Proceedings of the 24th European transport forum, PTRC

  25. Chang G-L, Su C-C (1995) Predicting intersection queue with neural network models. Transp Res Part C 3(3):175–191

    Article  Google Scholar 

  26. Zheng CJ, Lin DG, Zheng SK, Chen SY (2013) Research on traffic delay predicting at signalized intersection based on neural network. Appl Mech Mater 427–429:901–904. https://doi.org/10.4028/www.scientific.net/amm.427-429.901

    Article  Google Scholar 

  27. Ledoux C (1997) An urban traffic flow model integrating neural networks. Transp Res Part C Emerg Technol 5:287–300

    Article  Google Scholar 

  28. Gilmore JE, Abe N (1995) Neural network models for traffic control and congestion prediction. IVHS J 2(3):231–252

    Google Scholar 

  29. Huang PX, Xu G (1996) A neural network approach for predicting network traffic conditions. In: Proceedings of the 1996 annual meeting of ITS America, pp 507–516

  30. Dia H (2001) An object-oriented neutral network approach to short-term traffic forecasting. Eur J Oper Res 131:253–261

    Article  Google Scholar 

  31. Wang Y, Nihan NL (2000) Freeway traffic speed estimation with single-loop outputs. Transp Res Rec 1727:120–126

    Article  Google Scholar 

  32. Yin B, Dridi M, Moudni AE (2019) Recursive least-squares temporal difference learning for adaptive traffic signal control at intersection. Neural Comput Appl 31:1013–1028

    Article  Google Scholar 

  33. Williams BM, Durvasula PK, Brown DE (1998) Urban traffic flow prediction: application of seasonal autoregressive integrated moving average and exponential smoothing models. Transp Res Rec 1644:132–144

    Article  Google Scholar 

  34. Ahmed MS, Cook AR (1979) Analysis of freeway traffic time-series data by using Box–Jenkins techniques. Transp Res Board 722:1–9

    Google Scholar 

  35. Chung E, Rosalion N (2001) Short term traffic flow prediction. In: Proceedings of the 24th Australian transportation research forum, Hobart, Tasmania

  36. Stathopoulos A, Karlaftis MG (2003) A multivariate state-space approach for urban traffic flow modelling and prediction. Transp Res Part C 11(2):121–135

    Article  Google Scholar 

  37. Chen M, Chien SIJ (2001) Dynamic freeway travel time prediction using probe vehicle data: link-based versus path-based. In: Proceedings of the transportation research board 80th annual meeting, Washington, DC

  38. Clark S (2003) Traffic prediction using multivariate nonparametric regression. J Transp Eng 129(2):161–168

    Article  Google Scholar 

  39. Xu H, Jiang C (2020) Deep belief network-based support vector regression method for traffic flow forecasting. Neural Comput Appl 32:2027–2036

    Article  Google Scholar 

  40. Xiao Y, Yin Y (2019) Hybrid LSTM neural network for short-term traffic flow prediction. Information 10(3):105. https://doi.org/10.3390/info10030105

    Article  Google Scholar 

  41. Qiao S, Sun R, Fan G, Liu J (2017) Short-term traffic flow forecast based on parallel long short-term memory neural network. In: 2017 8th IEEE International conference on software engineering and service science (ICSESS). https://doi.org/10.1109/icsess.2017.8342908

  42. Li R, Lu H (2009) Combined neural network approach for short-term urban freeway traffic flow prediction. In: Advances in neural networks—ISNN 2009 lecture notes in computer science, pp 1017–1025. https://doi.org/10.1007/978-3-642-01513-7_112

  43. Castillo E, Sanchez-Cambronero S, Maria J (2010) Bayesian networks methods for traffic flow prediction. Bayesian Netw. https://doi.org/10.5772/10063

    Article  Google Scholar 

  44. Zheng W, Lee D-H, Shi Q (2006) Short-term freeway traffic flow prediction: bayesian combined neural network approach. J Transp Eng 132(2):114–121. https://doi.org/10.1061/(asce)0733-947x(2006)132:2(114)

    Article  Google Scholar 

  45. Ghofrani M, Carson D, Ghayekhloo M (2016) Hybrid clustering-time series-Bayesian neural network short-term load forecasting method. In: 2016 North American power symposium (NAPS). https://doi.org/10.1109/naps.2016.7747865

  46. Petridis V, Kehagias A, Petrou L, Bakirtzis A, Kiartzis S, Panagiotou H, Maslaris N (2001) A Bayesian multiple models combination method for time series prediction. J Intell Robot Syst 31(1/3):69–89. https://doi.org/10.1023/a:1012061814242

    Article  MATH  Google Scholar 

Download references

Acknowledgements

Many thanks go to Eng. Lulwa Alabdulmuhsen and Eng. Alaa Alsmadi for their efforts and supervision. We would also like to thank Eng. Sondos AlShimari, from the Central Department of Traffic at the Ministry of Interior, and all the people who provided us with the facilities being required and conductive conditions for our project.

Funding

Funding was provided by Kuwait University (Grant No. EV02/19).

Author information

Authors and Affiliations

  1. Civil Engineering Department, College of Engineering and Petroleum, Kuwait University, P.O. Box 5969, 13109, Safat, Kuwait, Kuwait

    Sharaf AlKheder, Wasan Alkhamees & Reyouf Almutairi

  2. Department of Mechanical Engineering, Abu Dhabi University, Abu Dhabi, United Arab Emirates

    Mohammad Alkhedher

Authors
  1. Sharaf AlKheder
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wasan Alkhamees
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Reyouf Almutairi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Mohammad Alkhedher
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Sharaf AlKheder.

Ethics declarations

Conflict of interest

We authors declare that we have no conflict of interest regarding our submitted article.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A: Data collection

See Fig. 22.

Fig. 22
figure 22

Data collection

Full size image

Appendix B

2.1 Cairo Street weekends model, true values

See Figs. 23, 24, 25 and 26.

Fig. 23
figure 23

Cairo Street weekends results model

Full size image
Fig. 24
figure 24

Time-series response (weekends/Cairo)

Full size image
Fig. 25
figure 25

Regression plot (weekends/Cairo)

Full size image
Fig. 26
figure 26

Summarized results for MSE and R values (weekends/Cairo)

Full size image

2.2 Cairo Street weekends model, normalized values

See Figs. 27, 28, 29 and 30.

Fig. 27
figure 27

Normalized Cairo Street weekends results model

Full size image
Fig. 28
figure 28

Normalized time-series response (weekends/Cairo)

Full size image
Fig. 29
figure 29

Normalized regression plot (weekends/Cairo)

Full size image
Fig. 30
figure 30

Summarized results for normalized MSE and R values (weekends/Cairo)

Full size image

Appendix C

3.1 Riyadh Street weekdays model, true values

See Figs. 31, 32, 33 and 34.

Fig. 31
figure 31

Riyadh Street weekdays results model

Full size image
Fig. 32
figure 32

Time-series response (weekdays/Riyadh)

Full size image
Fig. 33
figure 33

Regression plot (weekdays/Riyadh)

Full size image
Fig. 34
figure 34

Summarized results for MSE and R values (weekdays/Riyadh)

Full size image

3.2 Riyadh Street weekdays model, normalized values

See Figs. 35, 36, 37 and 38.

Fig. 35
figure 35

Normalized Riyadh Street weekdays results model

Full size image
Fig. 36
figure 36

Normalized time-series response (weekdays/Riyadh)

Full size image
Fig. 37
figure 37

Normalized regression plot (weekdays/Riyadh)

Full size image
Fig. 38
figure 38

Summarized results for normalized MSE and R values (weekdays/Riyadh)

Full size image

3.3 Riyadh Street weekends model, true values

See Figs. 39, 40, 41 and 42.

Fig. 39
figure 39

Riyadh Street weekends results model

Full size image
Fig. 40
figure 40

Time-series response (weekends/Riyadh)

Full size image
Fig. 41
figure 41

Regression plot (weekends/Riyadh)

Full size image
Fig. 42
figure 42

Summarized results for MSE and R values (weekends/Riyadh)

Full size image

3.4 Riyadh Street weekends model, normalized values

See Figs. 43, 44, 45 and 46.

Fig. 43
figure 43

Normalized Riyadh Street weekends results model

Full size image
Fig. 44
figure 44

Normalized time-series response (weekends/Riyadh)

Full size image
Fig. 45
figure 45

Normalized regression plot (weekends/Riyadh)

Full size image
Fig. 46
figure 46

Summarized results for normalized MSE and R values (weekends/Riyadh)

Full size image

Appendix D

4.1 Maghreb Street weekdays model, true values

See Figs. 47, 48, 49 and 50.

Fig. 47
figure 47

Maghreb Street weekdays results model

Full size image
Fig. 48
figure 48

Time-series response (weekdays/Maghreb)

Full size image
Fig. 49
figure 49

Regression plot (weekdays/Maghreb)

Full size image
Fig. 50
figure 50

Summarized results for MSE and R values (weekdays/Maghreb)

Full size image

4.2 Maghreb Street weekdays model, normalized values

See Figs. 51, 52, 53 and 54.

Fig. 51
figure 51

Normalized Maghreb Street weekdays results model

Full size image
Fig. 52
figure 52

Normalized time-series response (weekdays/Maghreb)

Full size image
Fig. 53
figure 53

Normalized regression plot (weekdays/Maghreb)

Full size image
Fig. 54
figure 54

Summarized results for normalized MSE and R values (weekdays/Maghreb)

Full size image

4.3 Maghreb Street weekends model, true values

See Figs. 55, 56, 57 and 58.

Fig. 55
figure 55

Maghreb Street weekends results model

Full size image
Fig. 56
figure 56

Time-series response (weekends/Maghreb)

Full size image
Fig. 57
figure 57

Regression plot (weekends/Maghreb)

Full size image
Fig. 58
figure 58

Summarized results for MSE and R values (weekends/Maghreb)

Full size image

4.4 Maghreb Street weekends model, normalized values

See Figs. 59, 60, 61 and 62.

Fig. 59
figure 59

Normalized Maghreb Street weekends results model

Full size image
Fig. 60
figure 60

Normalized time-series response (weekends/Maghreb)

Full size image
Fig. 61
figure 61

Normalized regression plot (weekends/Maghreb)

Full size image
Fig. 62
figure 62

Summarized results for normalized MSE and R values (weekends/Maghreb)

Full size image

Appendix E

5.1 Istiqlal Street weekdays model, true values

See Figs. 63, 64, 65 and 66.

Fig. 63
figure 63

Istiqlal Street weekdays results model

Full size image
Fig. 64
figure 64

Time-series response (weekdays/Istiqlal)

Full size image
Fig. 65
figure 65

Regression plot (weekdays/Istiqlal)

Full size image
Fig. 66
figure 66

Summarized results for MSE and R values (weekdays/Istiqlal)

Full size image

5.2 Istiqlal Street weekdays model, normalized values

See Figs. 67, 68, 69 and 70.

Fig. 67
figure 67

Normalized Istiqlal Street weekdays results model

Full size image
Fig. 68
figure 68

Normalized time-series response (weekdays/Istiqlal)

Full size image
Fig. 69
figure 69

Normalized regression plot (weekdays/Istiqlal)

Full size image
Fig. 70
figure 70

Summarized results for normalized MSE and R values (weekdays/Istiqlal)

Full size image

5.3 Istiqlal Street weekends model, true values

See Figs. 71, 72, 73 and 74.

Fig. 71
figure 71

Istiqlal Street weekends results model

Full size image
Fig. 72
figure 72

Time-series response (weekends/Istiqlal)

Full size image
Fig. 73
figure 73

Regression plot (weekends/Istiqlal)

Full size image
Fig. 74
figure 74

Summarized results for MSE and R values (weekends/Istiqlal)

Full size image

5.4 Istiqlal Street weekends model, normalized values

See Figs. 75, 76, 77 and 78.

Fig. 75
figure 75

Normalized Istiqlal Street weekdays results model

Full size image
Fig. 76
figure 76

Normalized time-series response (weekends/Istiqlal)

Full size image
Fig. 77
figure 77

Normalized regression plot (weekends/Istiqlal)

Full size image
Fig. 78
figure 78

Summarized results for normalized MSE and R values (weekends/Istiqlal)

Full size image

Appendix F

See Tables 4, 5, 6, 7 and 8.

Table 4 Actual traffic volume entering Cairo Street from Soor/Cairo intersection
Full size table
Table 5 Actual traffic volume entering Cairo Street from First Ring/Cairo intersection
Full size table
Table 6 Traffic volume forecasting of Cairo St. during weekdays
Full size table
Table 7 Normalized traffic volume data of Cairo St. during weekdays
Full size table
Table 8 Traffic volume results data of Cairo St. during weekends
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

AlKheder, S., Alkhamees, W., Almutairi, R. et al. Bayesian combined neural network for traffic volume short-term forecasting at adjacent intersections. Neural Comput & Applic 33, 1785–1836 (2021). https://doi.org/10.1007/s00521-020-05115-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-020-05115-y

Keywords

Search

Navigation