Estimating Traffic Accidents in Turkey Using Differential Evolution Algorithm
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study addresses the critical need for accurate traffic accident estimation to support road safety planning in Turkey, a country facing high accident rates due to concentrated highway transportation and rapid growth in population and vehicle ownership. The authors propose using the Differential Evolution Algorithm (DEA), a meta-heuristic optimization method, to develop predictive models for traffic accidents. The research aims to determine the most effective mathematical form for estimating accident numbers based on two primary parameters: population and the number of motor vehicles. The methodology utilizes historical data from the Turkish Statistical Institute covering the period from 2000 to 2014. The authors developed three distinct model forms—linear, exponential, and semi-quadratic—using DEA to optimize the weighting factors for each equation. The dataset was randomly split into training and testing subsets to verify model performance. The DEA process involved standard operators: initialization, mutation based on vector differences, crossover, and selection. The models were evaluated using statistical error criteria, specifically Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). The results indicate that while all three models successfully captured the increasing trend of traffic accidents, the linear model demonstrated superior performance. Statistical comparison revealed that the linear model had lower RMSE and MAPE values for both training and testing phases compared to the exponential and semi-quadratic forms. For instance, the linear model achieved a MAPE of 6.37% for training and 13.05% for testing, outperforming the other models. Consequently, the linear model was selected as the best fit for future predictions. Using this model, the authors projected accident numbers for the decade from 2015 to 2024, assuming population growth to approximately 85 million and vehicle ownership reaching 34 million. The forecast predicts a steady increase in accidents, rising from 1,415,052 in 2015 to 2,245,467 in 2024. The significance of this work lies in demonstrating the suitability of the Differential Evolution Algorithm for road safety applications and accident prediction. The study confirms that simple linear relationships between population, vehicle count, and accidents can provide reliable estimates when optimized via DEA. The findings underscore the urgent need for preventive traffic safety measures in Turkey, as the projected rise in accidents highlights a growing public health challenge. The authors suggest that future studies could incorporate additional parameters, such as road geometry and terrain conditions, to further enhance model accuracy.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | canonical_url | — | — | 1 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: crash risk outcomes