Home/Who it serves
Problem shapes
Four ways a week goes wrong.
These describe shapes of scheduling problem, not a client list. No customer work gets named on this site until there is a signed reference to name.
Field service
Engineers moving between sites all day, where travel time quietly decides whether the last job happens at all.
- Travel modelled between actual locations, not a flat allowance
- Skills and parts checked before a job gets assigned
- Customer promise times visible while the day gets planned
The plan gets drawn the night before. One late job at ten and the rest of the day gets rebuilt over the phone, by the person with the best memory.
A slip triggers a rebuild of what remains. The customers affected get told before they think to ring, and the reason for each move is recorded.
Appointments
Rooms, staff and equipment that all have to line up at the same moment, against a booking someone has already been promised.
- Room, person and kit held as one linked booking
- Cancellations offered to the waiting list at once
- Preparation and turnaround time counted as real work
Three diaries hold three versions of the truth. Gaps nobody notices sit next to double bookings nobody predicted.
One model holds every resource together, so a slot only exists when all of it is genuinely available.
Shift rotas
Coverage that has to hold across weeks, with leave, skills mix and working time rules all pulling against each other.
- Working time limits and rest periods applied, never assumed
- Skills mix checked on every shift, not just on paper
- Unpopular shifts spread evenly and the spread made visible
A spreadsheet only one person really understands, rebuilt each month, where fairness gets argued rather than measured.
Rules stated once and applied every cycle. The reasoning behind a rota can be inspected by anyone who asks.
Bays and workshops
Fixed capacity that will not stretch, where the binding constraint is a bay, a lift or a machine rather than a person.
- Physical capacity treated as the hard limit it actually is
- Queue position and promised dates linked to real throughput
- Servicing and calibration windows blocked well in advance
Work gets accepted on optimism, then queues in the yard while one bay absorbs every difficult job.
The model says no early, which costs far less than saying sorry late, and it shows the capacity that would change the answer.
Signals
Worth a call if any of these sound like the week.
- The schedule lives in one person's head, and holidays get planned around them
- A change at nine in the morning costs an hour of phone calls
- Nobody can say what today's capacity actually is until the day ends
- The plan looks fine and the team knows by lunchtime that it was not
- Work gets promised on optimism, then explained away afterwards
- Two teams keep separate diaries and both believe theirs is correct
None of these quite fit?
Scheduling problems rarely match a category cleanly. Describe the week and get a straight answer on whether a system is the right tool.