8.3 Crash Collision Typologies & Statistical Rate Formulations (MEV, MVMT)

8.3 Crash Collision Typologies & Statistical Rate Formulations (MEV, MVMT)

PTOE Exam Focus: Quantitative safety analysis problems on the PTOE examination heavily feature intersection crash rates ($R_{\text{MEV}}$), roadway segment crash rates ($R_{\text{MVMT}}$), the ANSI D16.1 KABCO injury scale, and the Critical Crash Rate statistical screening method ($R_c$). Candidates must be capable of computing exposure metrics, calculating rates with exact dimensional constants ($10^6$ vs. $10^8$), and determining statistical significance using Critical Rate Ratios (CRR).

---

1. Crash Collision Typologies & Injury Mechanisms

Traffic safety engineering classifies vehicular crashes based on the geometric orientation of colliding bodies, movement vectors, and object interaction:

+-----------------------------------------------------------------------------------+
|                         MAJOR CRASH COLLISION TYPOLOGIES                          |
|                                                                                   |
|  1. Angle / Broadside (T-Bone)    --> High severity at intersections / driveways  |
|  2. Rear-End                      --> High frequency at queues / signal approaches|
|  3. Head-On / Opposite Sideswipe  --> Severe kinetic energy on undivided roads    |
|  4. Single-Vehicle Roadway Dep.   --> Run-off-road, rollovers, fixed objects      |
|  5. Same-Direction Sideswipe      --> Weaving, merging, lane changes              |
|  6. Non-Motorist (Ped / Bike)     --> Vulnerable road user conflicts              |
+-----------------------------------------------------------------------------------+

• Angle / Broadside Collisions: Most frequent at unsignalized and signalized intersections where vehicle paths cross perpendicularly or obliquely. Angle crashes carry high fatality and injury severity because passenger vehicle side structures provide minimal energy-absorbing crush distance.
• Rear-End Collisions: Account for the highest raw frequency of crashes on signalized arterial corridors, freeway bottlenecks, and work zones. Primary contributing factors include tailgating, sudden queue deceleration, distracted driving, and inadequate perception-reaction buffers.
• Head-On Collisions: Characterized by massive kinetic energy dissipation ($E_k \propto (v_1 + v_2)^2$). While representing less than $3%$ of total national crashes, head-on collisions account for nearly $30%$ of fatalities on undivided two-lane rural highways.
• Single-Vehicle Roadway Departure (SVRD / ROR): Occur when a vehicle crosses an edge line or centerline, leading to rollovers or impacts with fixed roadside hazards (trees, utility poles, bridge abutments). SVRD crashes account for over $50%$ of all US highway fatalities.
• Same-Direction Sideswipes: Common in multilane weaving segments, interchange merge/diverge areas, and multilane roundabouts. Typically result in low-severity property damage unless secondary rollover occurs.
• Non-Motorist (Pedestrian & Cyclist) Crashes: Characterized by extreme vulnerability and high KABCO injury severity due to the complete absence of physical vehicle shielding.

---

2. The ANSI D16.1 KABCO Injury Severity Scale

The American National Standards Institute (ANSI D16.1) standardizes crash injury reporting across all US police accident reports and safety databases:

+-----------------------------------------------------------------------------+
|                          THE KABCO INJURY SCALE                             |
|                                                                             |
|   [ K ] Fatal Injury        --> Death within 30 days of the collision       |
|   [ A ] Suspected Serious   --> Incapacitating (broken bone, severe bleed)  |
|   [ B ] Suspected Minor     --> Evident non-incapacitating (abrasion, bruise)|
|   [ C ] Possible Injury     --> Non-visible complaint (dizziness, pain)     |
|   [ O ] Property Damage Only--> No injury; physical vehicle damage only     |
+-----------------------------------------------------------------------------+

• K — Fatal Injury: An injury that results in death within 30 days of the motor vehicle crash.
• A — Suspected Serious Injury (Incapacitating): Any injury other than fatal that prevents the injured person from walking, driving, or normally continuing the activities they were capable of performing before the crash (e.g., severe lacerations, broken limbs, skull fractures, abdominal injuries).
• B — Suspected Minor Injury (Non-Incapacitating): An evident injury that is not serious or fatal, such as minor abrasions, shallow cuts, minor bruising, or limping.
• C — Possible Injury: Any reported or claimed injury that is not evident or visible, such as momentary unconsciousness, claims of neck pain (whiplash), dizziness, or limping without visible damage.
• O — Property Damage Only (PDO / No Injury): A collision resulting strictly in property or vehicular damage with zero personal injuries.

---

3. Intersection Crash Rate Formulation ($R_{\text{MEV}}$)

Intersection safety performance is evaluated by normalizing total observed collisions against vehicular exposure measured in Million Entering Vehicles (MEV):

$R_{\text{MEV}} = \frac{C \times 10^6}{365 \times N \times \text{AADT}_{\text{entering}}}$

Where:

• $R_{\text{MEV}}$ = intersection crash rate (crashes per million entering vehicles, crashes/MEV)
• $C$ = total observed crashes at the intersection during the study period
• $N$ = duration of the study period in years (typically 3 to 5 years)
• $\text{AADT}_{\text{entering}}$ = total sum of Average Annual Daily Traffic entering from all intersection approaches (veh/day): $\text{AADT}_{\text{entering}} = \sum_{i=1}^{\text{approaches}} \text{AADT}_i$
• $10^6$ = conversion multiplier for million entering vehicles

---

4. Roadway Segment Crash Rate Formulation ($R_{\text{MVMT}}$)

Roadway segment safety is evaluated by normalizing collisions against vehicular travel exposure measured in 100 Million Vehicle-Miles of Travel ($100\text{ M VMT}$ or HMVM):

$R_{\text{MVMT}} = \frac{C \times 10^8}{365 \times N \times \text{AADT} \times L}$

Where:

• $R_{\text{MVMT}}$ = segment crash rate (crashes per 100 million vehicle-miles of travel, crashes/100 M VMT)
• $C$ = total observed crashes along the segment during the study period
• $N$ = duration of the study period in years
• $\text{AADT}$ = Average Annual Daily Traffic on the segment (veh/day)
• $L$ = length of the roadway segment in miles
• $10^8$ = conversion multiplier for $100\text{ million vehicle-miles}$ ($100 \times 10^6 = 10^8$)
• Total exposure: $\text{VMT}_{\text{total}} = 365 \times N \times \text{AADT} \times L$

---

5. The Critical Crash Rate Method

Comparing raw crash rates directly across sites can produce misleading conclusions because low-volume sites can exhibit artificially high rates due to random small-sample fluctuations. The Critical Crash Rate Method applies a Poisson-based statistical quality control model to establish a threshold rate above which a location is deemed statistically hazardous at a chosen confidence level:

$R_c = R_a + K \sqrt{\frac{R_a}{M}} + \frac{1}{2M}$

Where:

• $R_c$ = critical crash rate for the specific facility (crashes/MEV or crashes/100 M VMT)
• $R_a$ = average crash rate for all peer facilities of the same functional class
• $M$ = vehicular exposure of the facility during the study period:
  • For intersections: $M = \frac{365 \times N \times \text{AADT}_{\text{entering}}}{10^6}$ (in MEV)
  • For segments: $M = \frac{365 \times N \times \text{AADT} \times L}{10^8}$ (in 100 M VMT)
• $K$ = statistical confidence constant corresponding to the chosen level of significance:
  • $K = 1.282$ for $80%$ confidence ($\alpha = 0.20$)
  • $K = 1.645$ for $90%$ confidence ($\alpha = 0.10$)
  • $K = 1.960$ for $95%$ confidence ($\alpha = 0.05$, standard engineering default)
  • $K = 2.576$ for $99%$ confidence ($\alpha = 0.01$)
• $\frac{1}{2M}$ = continuity correction factor

Critical Rate Ratio (CRR):

$\text{CRR} = \frac{R_{\text{observed}}}{R_c}$

• If $\text{CRR} > 1.0$: The location's crash rate is statistically significantly higher than the peer group average, indicating a high-priority safety problem rather than random variation.
• If $\text{CRR} \le 1.0$: The observed crash frequency is within expected statistical limits.

---

6. Worked PTOE Engineering Calculation Examples

Example 1: Intersection Crash Rate ($R_{\text{MEV}}$)

Problem: A 4-leg suburban intersection recorded $C = 42$ crashes over a 3-year study period ($N = 3$). The 24-hour entering volumes are:

• Northbound Approach: $6,500\text{ veh/day}$
• Southbound Approach: $7,200\text{ veh/day}$
• Eastbound Approach: $4,800\text{ veh/day}$
• Westbound Approach: $5,500\text{ veh/day}$

Calculate the intersection crash rate in crashes per million entering vehicles ($R_{\text{MEV}}$).

Solution:

1. Sum total entering daily traffic: $\text{AADT}_{\text{entering}} = 6500 + 7200 + 4800 + 5500 = 24,000\text{ veh/day}$
2. Calculate total exposure over 3 years: $\text{Exposure} = 365 \times 3 \times 24,000 = 26,280,000\text{ entering vehicles} = 26.28\text{ MEV}$
3. Calculate intersection crash rate: $R_{\text{MEV}} = \frac{42 \times 10^6}{26,280,000} = \frac{42}{26.28} = 1.598 \approx 1.60\text{ crashes/MEV}$

Example 2: Segment Crash Rate ($R_{\text{MVMT}}$) & Critical Rate Screening

Problem: A $4.5\text{-mile}$ rural two-lane highway segment ($L = 4.5\text{ mi}$) has an AADT of $9,000\text{ veh/day}$. Over a 5-year period ($N = 5$), $92$ crashes occurred along the corridor. The statewide average crash rate for similar rural two-lane highways is $R_a = 95.0\text{ crashes per }100\text{ M VMT}$. Using a $95%$ confidence level ($K = 1.960$), determine whether this segment has a statistically significant crash rate.

Solution:

1. Compute segment vehicular exposure ($M$ in 100 M VMT): $M = \frac{365 \times 5 \times 9,000 \times 4.5}{10^8} = \frac{73,912,500}{10^8} = 0.7391\text{ HMVM}$
2. Compute observed crash rate ($R_{\text{observed}}$): $R_{\text{observed}} = \frac{92 \times 10^8}{73,912,500} = \frac{92}{0.7391} = 124.47\text{ crashes/100 M VMT}$
3. Calculate Critical Crash Rate ($R_c$): $R_c = R_a + K \sqrt{\frac{R_a}{M}} + \frac{1}{2M} = 95.0 + 1.960 \sqrt{\frac{95.0}{0.7391}} + \frac{1}{2(0.7391)}$ $\sqrt{\frac{95.0}{0.7391}} = \sqrt{128.535} = 11.3373$ $R_c = 95.0 + 1.960(11.3373) + 0.6765 = 95.0 + 22.221 + 0.6765 = 117.90\text{ crashes/100 M VMT}$
4. Evaluate Critical Rate Ratio (CRR): $\text{CRR} = \frac{R_{\text{observed}}}{R_c} = \frac{124.47}{117.90} = 1.056$ Because $\text{CRR} = 1.056 > 1.0$, the segment exhibits a statistically significant high crash rate at the $95%$ confidence level and warrants targeted engineering investigation.

ANSI D16.1 KABCO Scale Definitions and FHWA Comprehensive Injury Economic Costs (2024 USD)