# Identifying Hazardous Crash Locations Using Empirical Bayes and Spatial Autocorrelation

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Literature Review

Method | Advantage | Disadvantage | Suitable Condition | Related Work |
---|---|---|---|---|

Accident frequency | Considers the length and functionality of a road section. | Does not incorporate the regression effect of crashes. | Applicable to less traffic with a similar condition. | [3] |

Matrix analysis | Evaluation of results is flexible and accurate. | Subjective identification criteria. | Applicable to less traffic with a similar condition. | [23] |

Accident Severity | Considers types of crashes. | Inadequate representation of factors. | Applicable to a well-defined severity and consistent crash data. | [24] |

Accident rate | Considers many crash factors. | Needs a huge crash dataset and ignores the randomness of crash events. | Applicable to rural roads. | [4] |

Joint model (crash count and severity) | Considers correlated errors between crash count and severity. | Model complexity (Difficult to interpret and implement). | Applicable to different geographic scales. | [25] |

Equivalent accidents number | Considers many crash factors. | Needs a huge crash dataset and difficult to estimate the value of weight. | Applicable to urban roads with a similar condition. | [26] |

Quality control | Considers functionality of a road section and evaluation of result is accurate. | Needs huge traffic data and classification task. | Applicable to low traffic road sections. | [27] |

Accident spacing distribution | Looks at the distribution of crashes. | It can be affected by the scale of the analysis. | Applicable to areas where crashes are occurring as a result of environmental factors. | [28] |

Cumulative frequency | Uses many basic traffic data. | Does not consider the condition of a crash. | Applicable to crashes of varying conditions. | [7] |

Regression analysis | Considers different factors for crashes. | Needs huge basic data and many model parameters. | Applicable to the quantification of rural crashes. | [29] |

Fuzzy evaluation | Simple and suitable for multi-level problems. | Index of weight is subjective. | Wide applicability. | [30] |

Expert experience | Estimate a result easily and quickly. | It is quite subjective. | Applicable to roads that lack basic data. | [31] |

BP neural network | Evaluate crashes comprehensively. | An indicator is not directly related to a crash. | Applicable to highways. | [32] |

## 3. Materials and Methods

#### 3.1. Blackspot (BS) Analysis

#### 3.1.1. Preliminary Screening Based on Operational Definitions

#### 3.1.2. Final Segmentation Based on Unsupervised Machine Learning

#### **K-Means** **Clustering**

#### **Hierarchical** **Clustering**

#### 3.1.3. Empirical Bayes Method

- Determine whether the EB method is applicable,
- Determine whether observed crash frequency data are available,
- Assign crashes to individual roadway segments for use in the EB method,
- Apply the site-specific EB method.

#### 3.2. Optimised Hotspot Analysis (Getis-Ord Gi* Spatial Autocorrelation)

## 4. Results and Discussions

## 5. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Cheng, W.; Washington, S. Experimental evaluation of hotspot identification methods. Accid. Anal. Prev.
**2005**, 37, 870–881. [Google Scholar] [CrossRef] - Yuan, T.; Zeng, X.; Shi, T. Identifying Urban Road Black Spots with a Novel Method Based on the Firefly Clustering Algorithm and a Geographic Information System. Sustainability
**2020**, 12, 2091. [Google Scholar] [CrossRef][Green Version] - Johansson, Ö.; Wanvik, P.O.; Elvik, R. A new method for assessing the risk of accident associated with darkness. Accid. Anal. Prev.
**2009**, 41, 809–815. [Google Scholar] [CrossRef] - Weber, D.C. Accident Rate Potential: An Application of Multiple Regression Analysis of a Poisson Process. J. Am. Stat. Assoc.
**1971**, 66, 285–288. [Google Scholar] [CrossRef] - Da Costa, S.; Qu, X.; Parajuli, P.M. A Crash Severity-Based Black Spot Identification Model. J. Transp. Saf. Secur.
**2015**, 7, 268–277. [Google Scholar] [CrossRef] - Sugiyanto, G. The cost of traffic accident and equivalent accident number in developing countries (case study in Indone-sia). ARPN J. Eng. Appl. Sci.
**2017**, 12, 389–397. [Google Scholar] - Erdogan, S.; Yilmaz, I.; Baybura, T.; Gullu, M. Geographical information systems aided traffic accident analysis system case study: City of Afyonkarahisar. Accid. Anal. Prev.
**2008**, 40, 174–181. [Google Scholar] [CrossRef] [PubMed] - Pei, J.; Ding, J.; District, H. Improvement in the quality control method to distinguish the black spots of the road. J. Harbin Inst. Technol.
**2006**, 36, 97–100. [Google Scholar] - Erdogan, S. Explorative spatial analysis of traffic accident statistics and road mortality among the provinces of Turkey. J. Saf. Res.
**2009**, 40, 341–351. [Google Scholar] [CrossRef] - Joshua, S.C.; Garber, N.J. Estimating truck accident rate and involvements using linear and Poisson regression models. Transp. Plan. Technol.
**1990**, 15, 41–58. [Google Scholar] [CrossRef] - Yuntong, L. A fuzzy-based model for macroscopic evaluation of road traffic safety. China J. Highw. Transp.
**1995**, 8, 169–175. [Google Scholar] - MacNab, Y.C. A Bayesian hierarchical model for accident and injury surveillance. Accid. Anal. Prev.
**2003**, 35, 91–102. [Google Scholar] [CrossRef] [PubMed] - Chong, M.; Abraham, A.; Paprzycki, M. Traffic accident analysis using machine learning paradigms. Informatica
**2005**, 29, 89–98. [Google Scholar] - Qu, X.; Meng, Q. A note on hotspot identification for urban expressways. Saf. Sci.
**2014**, 66, 87–91. [Google Scholar] [CrossRef][Green Version] - Ord, J.K.; Getis, A. Local Spatial Autocorrelation Statistics: Distributional Issues and an Application. Geogr. Anal.
**2010**, 27, 286–306. [Google Scholar] [CrossRef] - OECD Road Research Group. Hazardous Road Locations–Identification and Counter Measures; Organisation for Economic Co-Operation and Development: Paris, France, 1976.
- Sørensen, M.W.J.; Elvik, R. Black Spot management and Safety Analysis of Road Networks: Best Practice Guidelines and Implementation Steps; Transportøkonomisk Institutt: Oslo, Norway, 2007. [Google Scholar]
- Ghadi, M.; Török, Á. A comparative analysis of black spot identification methods and road accident segmentation methods. Accid. Anal. Prev.
**2019**, 128, 1–7. [Google Scholar] [CrossRef] - Ghadi, M.; Török, Á.; Tánczos, K. Integration of Probability and Clustering Based Approaches in the Field of Black Spot Identification. Period. Polytech. Civ. Eng.
**2018**, 63, 46–52. [Google Scholar] [CrossRef] - Zou, Y.; Zhong, X.; Ash, J.; Zeng, Z.; Wang, Y.; Hao, Y.; Peng, Y. Developing a Clustering-Based Empirical Bayes Analysis Method for Hotspot Identification. J. Adv. Transp.
**2017**, 2017, 5230248. [Google Scholar] [CrossRef] - Wan, Y.; He, W.; Zhou, J. Urban Road Accident Black Spot Identification and Classification Approach: A Novel Grey Verhuls–Empirical Bayesian Combination Method. Sustainability
**2021**, 13, 11198. [Google Scholar] [CrossRef] - Ghadi, M.; Török, Á. Comparison Different Black Spot Identification Methods. Transp. Res. Procedia
**2017**, 27, 1105–1112. [Google Scholar] [CrossRef] - Dalai, B.; Landge, V.S. Risky zones in urban area: An analysis using fault tree and risk matrix method. Innov. Infrastruct. Solutions
**2022**, 7, 101. [Google Scholar] [CrossRef] - Hasan, M.S. Accident Analysis and Method Comparison of Finding Black Spots on M-2(Lahore-Islamabad) Motorway, Pakistan. Int. J. Res. Appl. Sci. Eng. Technol.
**2022**, 10, 25–38. [Google Scholar] [CrossRef] - Afghari, A.P.; Haque, M.; Washington, S. Applying a joint model of crash count and crash severity to identify road segments with high risk of fatal and serious injury crashes. Accid. Anal. Prev.
**2020**, 144, 105615. [Google Scholar] [CrossRef] [PubMed] - Sugiyanto, G.; Fadli, A.; Santi, M.Y. Identification of Black Spot and Equivalent Accident Number Using Upper Control Limit Method. ARPN J. Eng. Appl. Sci.
**2017**, 12, 528–535. [Google Scholar] - Dawei, X.; Xiansheng, L. Identification of Speedway Accident Black Spots Based on the Quality Control Method. In Proceedings of the 2015 Seventh International Conference on Measuring Technology and Mechatronics Automation, Nanchang, China, 13–14 June 2015; pp. 541–544. [Google Scholar] [CrossRef]
- Cui, H.; Dong, J.; Zhu, M.; Li, X.; Wang, Q. Identifying accident black spots based on the accident spacing distribution. J. Traffic Transp. Eng. (Engl. Ed.)
**2022**, 9, 1017–1026. [Google Scholar] [CrossRef] - Washington, S.; Haque, M.; Oh, J.; Lee, D. Applying quantile regression for modeling equivalent property damage only crashes to identify accident blackspots. Accid. Anal. Prev.
**2014**, 66, 136–146. [Google Scholar] [CrossRef][Green Version] - Yulong, P.; Han, G.; Tongyu, D. Fuzzy Evaluating Method to Distinguish the Black Spot of the Road; Institute of Transportation Research, Harbin Institute of Technology: Harbin, China, 2021; Available online: https://www.ictct.net/wp-content/uploads/V-Beijing-2007/ictct_document_nr_551_6_3PeiYulong221_230.pdf (accessed on 15 August 2021).
- Roudini, S.; Keymanesh, M.; Ahangar, A.N. Identification of “Black Spots” without Using Accident Information. Bull. Soc. Roy. Sc. Liège
**2017**, 667–676. [Google Scholar] [CrossRef] - Fan, Z.; Liu, C.; Cai, D.; Yue, S. Research on black spot identification of safety in urban traffic accidents based on machine learning method. Saf. Sci.
**2019**, 118, 607–616. [Google Scholar] [CrossRef] - Zhao, X.; Lord, D.; Peng, Y. Examining Network Segmentation for Traffic Safety Analysis With Data-Driven Spectral Analysis. IEEE Access
**2019**, 7, 120744–120757. [Google Scholar] [CrossRef] - Qin, X.; Wellner, A. Segment Length Impact on Highway Safety Screening Analysis. 2012. Available online: https://www.semanticscholar.org/paper/Segment-Length-Impact-on-Highway-Safety-Screening-Qin-Wellner/0595e2188788508e4ccf89b51d4916945ff1b1af (accessed on 23 March 2021).
- ArcMap 10.1.3 Support. 2019. Available online: https://desktop.arcgis.com/en/arcmap/10.3/guide-books/extensions/geostatistical-analyst/semivariogram-and-covariance-functions.htm (accessed on 15 August 2021).
- Sipos, T.; Mekonnen, A.A.; Szabó, Z. Spatial Econometric Analysis of Road Traffic Crashes. Sustainability
**2021**, 13, 2492. [Google Scholar] [CrossRef] - Dangeti, P. Statistics for Machine Learning: Techniques for Exploring Supervised, Unsupervised, and Reinforcement Learning Models with Python and R; Packt Publishing: Birmingham, UK, 2017. [Google Scholar]
- Forgy, E.W. Cluster analysis of multivariate data: Efficiency versus interpretability of classifications. Biometrics
**1965**, 21, 768–769. [Google Scholar] - Sazi, Y. Fuzzy Clustering Approach for Accident Black Spot Centers Determination. In Fuzzy Logic-Emerging Technologies and Applications; Dadios, E., Ed.; InTech: London, UK, 2012. [Google Scholar] [CrossRef][Green Version]
- Rousseeuw, P.J. Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. J. Comput. Appl. Math.
**1987**, 20, 53–65. [Google Scholar] [CrossRef][Green Version] - Calinski, T.; Harabasz, J. A dendrite method for cluster analysis. Commun. Stat.
**1974**, 3, 1–27. [Google Scholar] [CrossRef] - Davies, D.L.; Bouldin, D.W. A Cluster Separation Measure. IEEE Trans. Pattern Anal. Mach. Intell.
**1979**, PAMI-1, 224–227. [Google Scholar] [CrossRef] - Halkidi, M.; Batistakis, Y.; Vazirgiannis, M. On Clustering Validation Techniques. J. Intell. Inf. Syst.
**2001**, 17, 107–145. [Google Scholar] [CrossRef] - Hauer, E.; Harwood, D.W.; Council, F.M.; Griffith, M.S. Estimating Safety by the Empirical Bayes Method: A Tutorial. Transp. Res. Rec. J. Transp. Res. Board
**2002**, 1784, 126–131. [Google Scholar] [CrossRef] - FHWA. Interactive Highway Safety Design Model, U.S. Department of Transportation. 2021. Available online: https://highways.dot.gov/research/safety/interactive-highway-safety-design-model/interactive-highway-safety-design-model-ihsdm-overview (accessed on 5 October 2022).
- FHWA. Highway Safety Improvement Program, U.S. Department of Transportation. 2010. Available online: https://highways.dot.gov/safety/data-analysis-tools/rsdp/rsdp-tools/highway-safety-improvement-program-hsip-manual (accessed on 5 October 2022).
- AASHTO. Highway Safety Manual. American Association of State Highway and Transportation Officials. 2010. Available online: https://www.highwaysafetymanual.org/Pages/default.aspx (accessed on 22 October 2022).
- Getis, A. Reflections on spatial autocorrelation. Reg. Sci. Urban Econ.
**2007**, 37, 491–496. [Google Scholar] [CrossRef] - Morethingsjapanese.com. What Is Optimised Hotspot. 2020. Available online: https://morethingsjapanese.com/what-is-optimized-hot-spot-analysis/ (accessed on 14 December 2021).

**Table 2.**Comparing the two machine learning algorithms with three different criteria to find out the best.

Indices | Clustering Models | |
---|---|---|

K-Means | Hierarchical | |

Silhoutte score | 0.409 | 0.399 |

Davies–Bouldin index | 0.775 | 0.724 |

Calinski–Harabasz score | 105.012 | 93.541 |

Cluster ID | Length (km) | Average Daily Traffic | N_{E} − N_{P} | Label |
---|---|---|---|---|

1 | 5.42 | 13,157 | −7.916 | - |

2 | 5.21 | 9069 | −1.667 | - |

3 | 4.83 | 8915 | 1.837 | BS |

4 | 3.26 | 8569 | 0.744 | BS |

5 | 3.23 | 10,827 | −1.727 | - |

6 | 2.29 | 12,595 | −1.328 | - |

7 | 4.45 | 10,873 | −3.526 | - |

8 | 7.52 | 10,911 | −8.065 | - |

9 | 4.46 | 10,404 | 0.837 | BS |

10 | 2.58 | 15,369 | −4.742 | - |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Mekonnen, A.A.; Sipos, T.; Krizsik, N.
Identifying Hazardous Crash Locations Using Empirical Bayes and Spatial Autocorrelation. *ISPRS Int. J. Geo-Inf.* **2023**, *12*, 85.
https://doi.org/10.3390/ijgi12030085

**AMA Style**

Mekonnen AA, Sipos T, Krizsik N.
Identifying Hazardous Crash Locations Using Empirical Bayes and Spatial Autocorrelation. *ISPRS International Journal of Geo-Information*. 2023; 12(3):85.
https://doi.org/10.3390/ijgi12030085

**Chicago/Turabian Style**

Mekonnen, Anteneh Afework, Tibor Sipos, and Nóra Krizsik.
2023. "Identifying Hazardous Crash Locations Using Empirical Bayes and Spatial Autocorrelation" *ISPRS International Journal of Geo-Information* 12, no. 3: 85.
https://doi.org/10.3390/ijgi12030085