Standard summary set
(tables + one chart each): by year · severity (KABCO: K fatal, A incapacitating/serious, B non-incap., C possible, O PDO) · crash type (rear-end, angle, left-turn, sideswipe, ped, bike, fixed-object…) · lighting/weather/surface · time-of-day/day-of-week · high-frequency locations map. Always call out K+A separately — severity drives priority.
Crash rates
- Intersection (per MEV):
R = (C × 1,000,000) / (365 × N × V), V = total entering ADT. - Segment (per 100 MVMT):
R = (C × 100,000,000) / (365 × N × ADT × L_miles). - Compare to district/statewide averages for like facilities where available
#status/verify; say “above average” only with a comparator cited.
Pattern-finding
Cross-tab type × severity × location; read narratives for the K/A crashes (always); look for over-representation vs exposure, not raw counts.
Free companions
PEDSAFE / BIKESAFE — feed the ped/bike crash patterns this summary surfaces (crash type, location, conditions) into FHWA’s selection tools and get back matched countermeasures with case studies; the bridge from this note to Countermeasure Selection and CMFs.
HSM upgrade
For serious work, expected crashes via SPF+EB beats observed counts (regression-to-mean) — Highway Safety Manual (HSM).
Crash diagram
Schematic arrows per collision at each hot intersection; template exhibit; reviewers love them, juniors skip them.
Practice question
12 crashes at a 25k-entering-ADT intersection over 5 yrs — rate per MEV? (≈0.26)
QA/QC reminder
Totals across every table reconcile to the same N; K+A narrative review documented.