Articles

Working notes on schedule quality, update review and schedule risk, written for practising schedulers and project controls professionals. No fluff, no beginner content.

Schedule Quality

The DCMA 14-point check, one check at a time

Each check explained the way a reviewer actually uses it: what it measures, the standard threshold, why it exists, and where the threshold deserves an argument rather than obedience.

6 August 2026 · 12 minute read

Update Cycle

How to review a monthly schedule update

A working sequence in the order that finds problems fastest: data date discipline, statusing hygiene, silent re-plans, float erosion, critical path movement and compression.

6 August 2026 · 9 minute read

Tool Mechanics

Why the same schedule gives different dates in P6 and Microsoft Project

Day conversion, calendars, progress recalculation, duration types, float and constraint semantics: the engine-level differences that date the same network differently, plus the checklist to run before comparing.

6 August 2026 · 8 minute read

File Formats

What survives the export: XER, P6 XML and MPP compared

A field guide to the four exchange formats, what each one drops in transit, and which to request when baselines, calendars, costs or coding structures have to arrive intact.

6 August 2026 · 9 minute read

Schedule Quality

Negative float: what it actually means and who owns it

Trace it to its anchor before arguing about it: the three usual producers, why the tools report it differently, the three recovery exits, and the one unforgivable move.

6 August 2026 · 8 minute read

Update Cycle

Out-of-sequence progress: retained logic, progress override, and what your dates are really telling you

Two P6 users can publish different completion dates from the same file by one scheduling option. What OOS really is, per relationship type, and why the durable fix is in the logic.

6 August 2026 · 8 minute read

Schedule Risk

Choosing Monte Carlo uncertainty ranges when you have no data

Why blanket percentage spreads say nothing, how to range by work-type reference class, why correlation moves the S-curve more than any range does, and what to report from invented inputs.

6 August 2026 · 8 minute read

Governance

When re-baselining is legitimate, and when it is laundering

The four triggers that justify moving the datum, the embarrassment-driven reset that never does, and the governance that keeps re-baselining defensible.

6 August 2026 · 7 minute read

Schedule Quality

The 600-day test: the five-minute experiment that exposes a defended date

Add 600 working days to a critical task on a copy and recalculate. If the finish does not move with it, something other than the network is holding the date.

6 August 2026 · 5 minute read