Reliability engineering is a discipline based on the use of data and averages to make predictions about unit and system failures. The predictions made through this data analysis can be amazingly accurate and can save school bus maintenance managers money, time and frustration.

The data used to make these accurate failure predictions is probably already captured in your service records. The calculations can be as simple or as complex as the system that you wish to create or the failures that you wish to predict.

Here’s a simple example that uses Mean Time Between Failures (MTBF) and Mean Time to Restore (MTTR) to analyze the effectiveness of predictive replacement of certain parts or components. MTBF is the amount of time that you can expect on average to operate a unit without failure. MTTR is the time it takes on average to get a unit back on the road.

Real-world computations
A school district in Indiana covers 160 square miles with 275 active units. Each bus logs an average of 10,732 miles per year. The buses are rotated out of the fleet when they reach 10 years old. The average number of miles the district logs in a year is 275 x 10,732, or 2,951,300. The average service life of a school bus in this district is 10 x 10,732, or 107,320 miles.

According to a mechanic in the district, four of the failure modes experienced by the district are wipers, headlamps, fuel pumps and transmissions. On average, a wiper is changed once every 1,000 miles. Headlamps are also changed once every 1,000 miles. Fuel pumps fail about every 80,000 miles, and transmissions last about 100,000 miles. It takes about an hour to change a headlamp or a wiper if we round up. It takes about eight hours to change a fuel pump and three work days to change out or rebuild a transmission.

Applying the numbers
Preventive maintenance calculations may prove that it is more cost effective to replace fuel pumps at 60,000 miles as a matter of course than to let one fail, leaving the vehicle and its occupants stranded on the road.

A thorough review of the fleet maintenance records will determine the optimum preventive maintenance schedule. The cost of the fuel pump will be the same whether it is a scheduled repair or an emergency repair. The retrieval cost of a bus on the road and the dissatisfaction of the occupants will be avoided if the fuel pump is replaced in a controlled environment (preventive maintenance) prior to an unscheduled catastrophic failure.

Transmissions fail when they may, and the cost of replacing transmissions toward the end of the useful life of the bus as a matter of preventive maintenance does not make sense from a value perspective. Recall that average bus life is 107,320. Transmissions may fail around 100,000. While preventive maintenance for fuel pumps appears to make financial sense, replacing a transmission prior to failure does not.

Wipers are changed during the 2,000-mile checkup or as required. Wipers are a minor repair item and should continue to be checked during routine bus checks. Headlamps are replaced when damaged, and there is no prediction for damage. Headlamps are checked during the 2,000- and 6,000-mile checkups and replaced as required.

Strategy should be developed
An in-depth review of the maintenance records along with the development of a replacement strategy based on the repairs for the fleet could prove to be a valuable undertaking. A preventive maintenance/forecasting model could be developed using existing data.

Several reliability computer programs are already available for fleet use. The American Society for Quality also certifies reliability engineers, so a garage can certify its own people to do these calculations and change out high-risk items prior to their predicted failure.

Dollar bills are waiting to be picked up all around the garage just by applying some simple reliability calculations to your fleet.

Ted Mattis is the quality manager at Honeywell Aerospace in South Bend, Ind. He is currently enrolled in a doctorate program in technology management at Indiana State University.