Fable
The first new carriage ran only the shortest delivery route
The Luyuan caravan had a batch of old horse carts. They ran slowly and broke down predictably. The drivers knew which stretches of road would be bumpy and which post stations could repair wheels.
The carriage workshop built a new carriage, fitted with automatic brakes and a route puppet. In-yard tests were beautiful, and the workshop wanted to immediately replace all the old carriages. The old coachman reminded them that real roads have mud pits, spot checks, sharp turns, and rule-breaking vendors. The question for the new carriage is not whether it can run, but how much cargo will be damaged when something goes wrong on a real road.
The caravan leader’s naive fix was to have an old driver accompany each new vehicle. But if all routes switched vehicles at the same time, the old drivers could not handle all anomalies simultaneously. So they deployed only the first new vehicle on the shortest route with the lowest cargo value and the most familiar customers. The vehicle carried extra recorders, and any braking, detours, delays, and manual takeovers were reported.
In the first week, the new vehicles frequently decelerated on slopes after rain. The problem did not spread across the whole network; based on the records, the vehicle workshop adjusted the braking threshold and added slope testing. In the second week, the new vehicles expanded to three similar routes. Only after the metrics stabilized did the caravan switch more routes over; high-value cold-chain cargo always waited until the very end.
This deployment approach let the carriage workshop see the real environment while the caravan kept control. Every expansion had data behind it, every pause had a reason, and every rollback had an old carriage to fall back on.
Heeding the advice, the carriage workshop let only the first new carriage run the shortest route with the lightest cargo, and arranged for an old carriage to follow behind. At every stop, the driver noted how the brakes, wheels, and cargo box responded. The first trip exposed small problems: after rain the wooden wheels slipped, and the route puppet made its judgment too late at the market entrance. If the whole fleet had been replaced, the loss would have been great; now only one carriage was delayed.
This small incident made everyone stop and review. They did not rush to swap in a more impressive-sounding explanation, but instead wrote down clearly, one by one, the entry points that went wrong, the rules that were misunderstood, and the boundaries that needed to be preserved.
Only after the repairs did the carriage workshop expand to three lines. The new carriage was not unusable; it just had to reveal its flaws on real roads in a small scope first. The caravan leader admitted that the new carriage would eventually become widespread, but the first step had to make the risk small enough to learn from, not so large that all you could do was apologize.