Every breakdown gets a line: which machine, how long it was down, what actually broke, what the fix cost, who fixed it - and the column that pays for the whole page: was it preventable? Written the hour the machine comes back, triaged fifteen minutes on Friday, rolled up once a month. That is the entire system, and it is the difference between a shop where breakdowns are weather and a shop where breakdowns are a line item you are slowly deleting.
The seven columns, in order: machine (named as the floor names it, plus the asset number); started/ended and the clock that matters - lost production, not wrench time (a 20-minute fix mid-run on your only oven can cost more than an 8-hour fix on an idle Sunday); what broke in plain words ("door gasket torn, coil iced, compressor short-cycling," never "machine down"); the fix and the part that ended it - this column is where your spares list shops; the cost (parts + outside labour + missed production); who fixed it; and preventable Y/N plus one line on what would have caught it.
Friday, fifteen minutes, three decisions. Every preventable = Y event becomes a candidate task on the preventive maintenance schedule. Any failure mode that has now appeared twice in ninety days jumps the queue - repeat offenders are the machines writing your future. Any fix that needed a part goes onto the critical spares list with the downtime tolerance attached. A log nobody reads is a diary; the triage is the product.
The monthly roll-up is where money appears. Sum hours and cost per machine and two things fall out. First, the 80/20 line: two machines usually hold most of the pain, and the dramatic one is often not either of them - the quiet leaker stopping twelve minutes twice a week outranks the four-hour Saturday. Second, the repair-or-replace threshold: when a rolling year of downtime cost passes roughly a third of replacement cost and the trend is not bending after PM and spares are in place, the machine goes on the capital list - and the budget gets a line item instead of a surprise.
Five traps: logging only the big outages (the fifteen-minute stops are where the money leaks); blame entries (who fixed it, never who broke it - a log that names culprits teaches the floor to log nothing); repair-hours-only accounting (carry both clocks, cost the lost-production one); the log nobody reads (the triage is the product, the entries are just raw material); and shoebox history (one page per machine per quarter beats a year of texts - the log must survive the person keeping it in his head).
The worked example: a 40-seat restaurant logged every kitchen stop for a quarter. Q1: six downtime events, 19.5 lost-production hours, $1,150 in repairs and call-outs - and about $7,300 of revenue walking out the door once missed covers were priced. The instinct was to blame the Saturday the combi oven threw an error. The roll-up said otherwise: four of the six events were the same walk-in cooler, same failure mode - compressor short-cycling on overload. Root cause: a torn $45 door gasket letting humid air in, icing the coil. The fix: a gasket, a defrost check added to the weekly cold-chain walk, a spare gasket on the shelf with the fiche in the drawer. Q2: one event, 2.5 hours, zero call-outs. The same roll-up put the 12-year-old combi oven on the capital list - $2,900/year in downtime suddenly had a replacement business case, a date, and a budget line. Nothing changed except that the breakdowns were being counted, and the counting did the arguing.
The full page - the seven columns in detail, the Friday triage, the monthly roll-up, the lemon list, and the worked example - lives at Hive Ops Notes.
Free: The First 30 Minutes - the incident quick-start checklist. If your operation needs the full system: Ops Starter Kit ($14), Ops Starter Kit Vol. 2 ($27), or the Ops Mega Bundle (all 5 kits, $49).
Top comments (0)