15 Transit KPI Interview Questions (with Answers)
Transport planner and service planner interviews almost always include a round on bus performance indicators. The interviewer wants to know two things: can you calculate them, and do you know when they lie? Below are the questions that come up most, with answers you can adapt. Practise saying them out loud.
The example we’ll use throughout — one route, one weekday (illustrative):
22 km each way · 36 one-way trips · 5 buses · 42 seats per bus · 1,800 passengers · ₹21,600 fare revenue · 9,900 passenger-km
Warm-up: definitions
1. What is earnings per kilometre (EPKM)?
Answer: Fare revenue divided by the kilometres the buses ran. Here, 22 km × 36 trips = 792 bus-km, so ₹21,600 ÷ 792 = ₹27.27 per km.
Likely follow-up: “Which kilometres?” Say whether you mean route length × trips, scheduled kilometres, or GPS kilometres — they give different answers.
2. And load factor?
Answer: How much of the seating on offer was used, measured in passenger-km. Seat-km = 792 × 42 = 33,264. Load factor = 9,900 ÷ 33,264 = 29.8%.
3. How is EPKM different from earnings per bus?
Answer: Earnings per bus divides revenue by the buses in service that day: ₹21,600 ÷ 5 = ₹4,320. A bus on short, busy trips can earn a lot per day and still have a modest EPKM.
4. What does average trip length tell you?
Answer: Passenger-km ÷ passengers = 9,900 ÷ 1,800 = 5.5 km. It describes how the city travels more than how well the route performs — useful for fare stages, not for judging a route.
5. How would you measure frequency from ticket records?
Answer: For each route and direction on a day: time from first to last departure, divided by (departures − 1). With 18 departures per direction between 06:30 and 20:00, that’s 810 ÷ 17 ≈ 48 minutes. Never combine different routes into one span — parallel routes would look like one very frequent service.
Spot the error
6. A colleague averaged the daily load factors for the month. Problem?
Answer: Yes — quiet days count as much as busy ones. Route A uses 1,800 of 3,000 seat-km (60%), route B 200 of 2,000 (10%). The average of the two percentages is 35%; the real combined figure is 2,000 ÷ 5,000 = 40%. Add up first, divide last.
7. Monthly bus-km came out roughly 30 times too high. Why?
Answer: Someone probably summed the route length over all daily rows (length × 30 days) and then multiplied by total trips for the month. Work out each day’s bus-km first (length × that day’s trips), then add the days up.
8. One route shows negative earnings per km. What happened?
Answer: Refunds or cancellations were added to sales and outweighed them, or the route’s length is recorded as negative. Keep refunds separate and reject any length of zero or less.
9. Load factor shows 0% for the whole city. What do you check?
Answer: Whether the ticket records carry a travel distance at all. If they don’t, passenger-km is unknown, not zero. I’d report “not available” and how many tickets lack a distance.
10. A report says city buses average 55 km/h. Do you believe it?
Answer: No — mixed city traffic with stops rarely allows anything close. The usual cause is treating the last ticket sold as the end of the trip, which makes every trip look short. I’d ask for recorded trip end times or GPS.
Applying it
11. How would you raise this route’s EPKM?
Answer: EPKM = revenue per passenger × passengers per bus-km. Here, ₹12 × 2.27 = ₹27.27. So I can either carry more people on the same kilometres — move trips from quiet hours to the peaks, turn some buses back before an empty outer section — or earn more per passenger, by checking for unissued tickets and fare stages that cut journeys short. I wouldn’t cut a route that exists for access without saying so.
12. Buses are full but EPKM is low. What’s going on?
Answer: Lots of riders paying little per kilometre: very short trips on a flat minimum fare, a high share of concessions or passes, or tickets not being issued. I’d compare revenue per passenger-km with similar routes and across conductors.
13. What can’t ticket records tell you?
Answer: Buses that ran without selling a single ticket, empty running to and from the depot, punctuality against a timetable, operating cost, transfers, and fare evasion.
14. If you also had GPS data, what would you add?
Answer: Real kilometres run, including buses that carried no paying passengers; empty running; true departure and arrival times, so speed and running time; and bunching. First I’d check what share of bus-days actually have GPS.
15. You get a new month of ticket data. What do you check first?
Answer: Rows per day against last month; duplicates; refunds; tickets whose route, stop or bus isn’t in the reference tables; route codes written different ways; and extremes in trips per bus and passengers per trip, which usually point to broken trip numbering.
Before the interview
- Work the example above on paper until each answer takes you under a minute.
- Prepare one story where you found an error in someone’s numbers and what it changed.
- Brush up on the wider skill set in transit planner skills for bus ticket data.