Subject: Re: Start and stop times -- what our own reports already record
Attachment: late-start-onepager.pdf

Suggestion: copy everyone who received the earlier overview. Tom raised
project sustainability, which makes this a group topic rather than a
one-to-one, and it saves having the same exchange four more times.

---


Hi Tom,

Following up on my earlier reply. Rather than wait, I went back through what
we already hold.

"Late Start / Early Finish (Hrs)" is not a concept I introduced for this
procedure. It is a standing column in all four of our productivity report
templates, filled in shift by shift at the quay. Across 1,267 reports from
2024 to 2026 that gives us 16,460 crane-shift records.

They show a late start or early finish in 55% of all vessel-shifts, totalling
1,658 crane-hours -- the equivalent of 207 full crane shifts, and 12% of all
the delay time we log. On break bulk it is the second largest delay cause; on
RoRo the largest. It is flat across day, afternoon and night, so it is not one
crew. The one-page breakdown is attached, including what the figures do not
prove.

So we were already recording it, and we already knew the pattern. What we have
never had is where and why, case by case. That is the point of the procedure:
it separates late start from early finish, puts a reason code against each
gap, and tells us whether the cause is something we can actually fix. Today we
can count the hours. We cannot act on them.

Worth being honest about the other half of it: we see it on the quay as well,
and we as GSVs mostly let it pass. I include myself in that. Not out of
unwillingness -- nothing is fed back, so nobody has to say anything, and a
late start leaves no trace by the end of the shift.

I would also note that this is the largest delay category sitting entirely
within our own control. "Delay due to CSO" is bigger, but it needs a
counterparty to resolve. This one does not.

So my proposal is to start recording now and bring the first weeks of live
data to the 21st. Then the meeting is not about whether the problem is real.
It is about what the numbers tell us to change, and whether the reason codes
match what the clerks actually see -- which we can only find out by running
it.

Kind regards,
Leon