Alberta Traffic Collision Analysis (2022)

Title Page

Alberta Traffic Collision Analysis (2022)
DAT 301 - Project 1
Author: Rudy
Date: 2025-03-27

Table of Contents

(Automatically generated)

1. Introduction

This report analyzes monthly traffic collision data from Alberta in 2022. The objective is to discover trends and draw conclusions that can support future safety measures across Alberta’s highways and city streets.

Traffic accidents are influenced by numerous variables, including weather, driver behavior, enforcement intensity, and traffic volume. Alberta presents a unique case due to its geographic and climate diversity—ranging from snowy winters to busy summer travel seasons.

Understanding the monthly and seasonal distribution of traffic collisions can help policymakers implement timely interventions and road safety campaigns. This project uses a manually extracted subset of the provincial traffic data to demonstrate how data analysis can support real-world decision-making.

2. Data Source and Description

The dataset was manually constructed using figures from the 2022 Alberta Traffic Collision Statistics PDF available on the Alberta Open Government Portal.

collision_data <- read_csv("C:\\Users\\iamru\\OneDrive\\Documents\\dat301\\Alberta_Collision_Monthly_2022.csv")

## Rows: 12 Columns: 4
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): Month
## dbl (3): Fatal_Collisions, Major_Injury_Collisions, Total_Collisions
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

glimpse(collision_data)

## Rows: 12
## Columns: 4
## $ Month                   <chr> "January", "February", "March", "April", "May"…
## $ Fatal_Collisions        <dbl> 14, 14, 8, 14, 26, 23, 36, 32, 29, 25, 18, 12
## $ Major_Injury_Collisions <dbl> 134, 112, 116, 114, 163, 184, 225, 242, 214, 2…
## $ Total_Collisions        <dbl> 10759, 9120, 8588, 7576, 7874, 8987, 8996, 881…

2.1 Column Descriptions

• Month: The calendar month
• Fatal_Collisions: Collisions that resulted in at least one fatality
• Major_Injury_Collisions: Serious injury collisions that led to hospitalization
• Total_Collisions: All recorded collisions (including minor and property damage)

summary(collision_data)

##     Month           Fatal_Collisions Major_Injury_Collisions Total_Collisions
##  Length:12          Min.   : 8.00    Min.   :112.0           Min.   : 7576   
##  Class :character   1st Qu.:14.00    1st Qu.:129.5           1st Qu.: 8760   
##  Mode  :character   Median :20.50    Median :168.5           Median : 8992   
##                     Mean   :20.92    Mean   :169.2           Mean   : 9743   
##                     3rd Qu.:26.75    3rd Qu.:214.8           3rd Qu.: 9677   
##                     Max.   :36.00    Max.   :242.0           Max.   :15034

skim(collision_data)

Data summary

Name	collision_data
Number of rows	12
Number of columns	4
_______________________
Column type frequency:
character	1
numeric	3
________________________
Group variables	None

Variable type: character

skim_variable	n_missing	complete_rate	min	max	empty	n_unique	whitespace
Month	0	1	3	9	0	12	0

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
Fatal_Collisions	0	1	20.92	8.85	8	14.00	20.5	26.75	36	▃▇▂▆▃
Major_Injury_Collisions	0	1	169.17	47.24	112	129.50	168.5	214.75	242	▇▂▃▂▅
Total_Collisions	0	1	9743.42	2165.73	7576	8759.75	8991.5	9676.75	15034	▇▂▁▁▁

3. Monthly Breakdown and Visualizations

collision_data$Month <- factor(collision_data$Month, levels = month.name)

3.1 Total Collisions by Month

ggplot(collision_data, aes(x = Month, y = Total_Collisions)) +
  geom_bar(stat = "identity", fill = "steelblue") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Total Traffic Collisions by Month", y = "Number of Collisions", x = "Month")

Interpretation:
December reported the highest number of traffic collisions (15,034), likely due to challenging winter driving conditions. January and November also show elevated collision counts, possibly related to icy roads and holiday travel. The lowest month was April, indicating more favorable driving conditions and fewer disruptions.

3.2 Fatal vs Major Injury Collisions

collision_data_long <- collision_data %>%
  pivot_longer(cols = c(Fatal_Collisions, Major_Injury_Collisions), names_to = "Type", values_to = "Count")

ggplot(collision_data_long, aes(x = Month, y = Count, group = Type, color = Type)) +
  geom_line(linewidth = 1.2) +
  geom_point(size = 2) +
  labs(title = "Fatal vs Major Injury Collisions by Month", x = "Month", y = "Count") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Interpretation:
July had the most fatal collisions (36), while August had the most major injuries (242). These summer peaks might be attributed to increased traffic during vacation season, more vulnerable road users (pedestrians, cyclists, motorcycles), and possibly driver fatigue on long trips.

4. Seasonal Trends

collision_data <- collision_data %>%
  mutate(Season = case_when(
    Month %in% c("December", "January", "February") ~ "Winter",
    Month %in% c("March", "April", "May") ~ "Spring",
    Month %in% c("June", "July", "August") ~ "Summer",
    TRUE ~ "Fall"
  ))

seasonal_summary <- collision_data %>%
  group_by(Season) %>%
  summarise(across(Fatal_Collisions:Total_Collisions, sum))

kable(seasonal_summary)

Season	Fatal_Collisions	Major_Injury_Collisions	Total_Collisions
Fall	72	566	31170
Spring	48	393	24038
Summer	91	651	26800
Winter	40	420	34913

Interpretation:
- Winter leads in total collisions due to slippery road conditions.
- Summer leads in fatalities and injuries—likely due to higher speeds and volume.

5. Boxplots by Season

collision_data %>%
  pivot_longer(cols = c(Fatal_Collisions, Major_Injury_Collisions), names_to = "Type", values_to = "Count") %>%
  ggplot(aes(x = Season, y = Count, fill = Type)) +
  geom_boxplot() +
  labs(title = "Collision Severity by Season")

Interpretation:
Boxplots illustrate how summer collisions tend to have a higher range and median for both fatalities and injuries, reinforcing the seasonal risks.

6. Percentage Change Month to Month

collision_data <- collision_data %>%
  arrange(match(Month, month.name)) %>%
  mutate(Pct_Change = round(100 * (Total_Collisions - lag(Total_Collisions)) / lag(Total_Collisions), 1))

kable(collision_data[, c("Month", "Total_Collisions", "Pct_Change")])

Month	Total_Collisions	Pct_Change
January	10759	NA
February	9120	-15.2
March	8588	-5.8
April	7576	-11.8
May	7874	3.9
June	8987	14.1
July	8996	0.1
August	8817	-2.0
September	8967	1.7
October	9316	3.9
November	12887	38.3
December	15034	16.7

Interpretation:
Significant jumps are seen in the lead-up to winter months and just after summer holidays. This can help predict when to ramp up road safety messaging.

7. Statistical Measures

collision_data %>%
  summarise(
    Mean_Fatal = mean(Fatal_Collisions),
    Median_Fatal = median(Fatal_Collisions),
    Max_Total = max(Total_Collisions),
    Min_Total = min(Total_Collisions)
  )

Interpretation:
The average fatal collisions per month was ~21, with July being well above the average. Understanding these ranges helps contextualize what counts as an “exceptional” month.

8. Context: Road Safety and Driver Behavior

Alberta’s traffic patterns are influenced by multiple factors. Research from Transport Canada and the Alberta Ministry of Transportation has shown that:

• The majority of fatal collisions involve driver error
• Distracted and impaired driving peak during weekends and holidays
• Use of winter tires reduces crash risk by 20–25% in icy conditions

Behavioral trends such as texting while driving, fatigue, and seasonal inexperience (e.g., summer drivers in rural areas) must be addressed in tandem with enforcement.

Alberta also sees a wide variety of road conditions across urban and rural areas. City driving introduces congestion and pedestrian risk, while highways present speeding and fatigue hazards. Understanding these environments is key for tailoring enforcement and safety campaigns.

9. Policy Implications and Practical Interventions

To reduce fatalities and injuries, Alberta’s municipalities and transport authorities can consider:

• Seasonal alerts and variable speed zones to adapt to ice and fog conditions
• Increased DUI checkpoints in July–August and December, when rates of impaired driving spike
• Mandatory winter tire regulations similar to B.C., requiring use during designated months
• Data-driven patrol deployment, focusing police presence during peak months identified by analysis

Infrastructure improvements could include smart signage, rumble strips at rural intersections, and improved lighting in high-collision areas. Public campaigns on distracted driving, speeding, and vehicle maintenance should align with these monthly trends.

10. Future Work and Recommendations

To build on this exploratory analysis, future projects could:

• Integrate GIS data to produce maps of high-risk collision zones across Alberta
• Analyze multi-year datasets to identify long-term trends or the effect of safety interventions
• Use machine learning to predict collision likelihood based on month, weather, and location
• Merge with hospital data to study injury outcomes and response times

Additionally, deeper behavioral data (e.g., seatbelt use, texting, alcohol) could be gathered from law enforcement databases to enhance predictive models and personalize campaigns. A more automated data pipeline would also allow near real-time dashboards for transport agencies.

11. Conclusion

This report successfully uncovered seasonal and monthly trends in Alberta’s 2022 collision data. It reinforces the importance of proactive planning and seasonal preparation.

With access to a larger and more granular dataset, the analysis could be expanded to include predictive modeling and spatial clustering.

12. References

• Alberta Open Data Portal (2022). Traffic Collision Statistics.
  https://open.alberta.ca/publications/traffic-collision-statistics

Appendix A: Full Dataset

kable(collision_data[, 1:4])

Month	Fatal_Collisions	Major_Injury_Collisions	Total_Collisions
January	14	134	10759
February	14	112	9120
March	8	116	8588
April	14	114	7576
May	26	163	7874
June	23	184	8987
July	36	225	8996
August	32	242	8817
September	29	214	8967
October	25	217	9316
November	18	135	12887
December	12	174	15034

Appendix B: Additional Figures

ggplot(collision_data, aes(x = Fatal_Collisions)) +
  geom_histogram(binwidth = 2, fill = "darkred", color = "white") +
  labs(title = "Histogram of Fatal Collisions", x = "Fatal Collisions", y = "Frequency")

ggplot(collision_data, aes(x = Total_Collisions)) +
  geom_density(fill = "lightblue") +
  labs(title = "Density Plot of Total Collisions")