Construction Sequencing and Method

Last reviewed 9 September 20262,445 words11 min read

Turning a method statement into logic — the sequence on paper becomes the links in P6, and each link needs a reason

🟩 In one line: A method statement describes how the work will be done; the schedule is that description translated into activities and relationships — one activity per step per zone, hard logic where physics demands it, soft logic where the method chooses it, each soft link labelled — so that when the method changes, the logic changes with a record, and when the logic is challenged, the method statement is the answer.

👤 Who this is for: J/M/S. Prerequisites: standard-sequences, hard-logic-vs-soft-logic, relationship-types-explained, crew-and-manhour-formulas.

e / c
First, let's be honest about why this page exists

The method statement is written by the construction manager, the schedule by the planner, often in different weeks and rarely from the same document. The result is a schedule whose logic nobody can trace to a method — 40 FS links on a floor where the method statement describes zoned, overlapping work — and a method statement that promises a sequence the schedule doesn't show. The first reviewer who asks "why does plaster wait for the whole floor's blockwork?" gets no answer.

The translation is mechanical once the habit exists: read the method, list the steps, name the zones, decide for every link whether the physics or the method put it there, and write that decision down. This page is that habit.

🟨 The standard — what "good" looks like
SourceWhat it asks for (paraphrased)Consequence
FIDIC 1999 Cl 8.3 / 2017 Cl 8.3Programme accompanied by a supporting report describing the methods the Contractor intends to adoptThe method statement and the logic are one submission — they must agree
NEC3/NEC4 Cl 31.2Programme shows order and timing, and the method statement for each operation, with resourcesSame, expressed as mandatory programme content
SCL Protocol 2nd ed., Core Principle 1 and Guidance Part BProgramme logic should reflect the intended sequence; records of changes to intent keptEvery logic change tied to a method change is recorded
AACE RP 24R-03 (Developing Activity Logic)Logic based on physical requirements, resource availability and preference; preferential logic identifiedThe hard/soft/resource tagging
AACE RP 38R-06Basis records sequencing assumptions and methodSBM section on sequence per area
GAO Schedule Assessment Guide, BP 2Sequence all activities with logic that reflects how the work will be performedSame
DCMA 14-Point, Metrics 1–4Logic present; leads zero; lags ≤ 5%; FS ≥ 90%The translation must produce mostly FS links with real lags only
Convention (hard-logic-vs-soft-logic)Test: with unlimited resources, would the link still be needed? Soft links labelled in relationship Comments or Notebook "Logic Basis"Applied to every link written

🟢 Rule: one activity per method step per zone; every relationship classified hard, soft-resource or soft-preference at the moment it is written and labelled if soft; overlaps modelled by zoning first and SS/FF pairs second, never by negative lag; hold points and inspections as activities; and the method statement revision number in the SBM next to the sequence it produced.

How it actually works

1. Read the method statement for six things

ExtractBecomes in P6Example
Steps, in orderActivities (verb + object + location)"Fix rebar L5 slab Zone A"
Zones, pours, bays, liftsThe zone code and the activity splitSlab in 2 pours → 2 activity chains
Temporary works and plantActivities (erect/strike) and LOE (crane, hoist)"Erect back-propping L4", "Tower crane TC1" LOE
Resources per stepCrew size in the duration basis UDF; resource logic where crews or sets are shared2 formwork sets → set returns after strike
Hold points, inspections, tests, curingActivities on the right calendar"Cure L5 slab Zone A" 7-day; "Pre-pour IR L5-A" 1 day
Access and interfacesMilestones and Start On or After with references"Podium roof released to façade" milestone

If the method statement doesn't say how many zones, crews or sets, ask before you draw a link. Every one of those numbers is logic.

2. Name the zones before you draw anything

Overlap between trades is modelled by splitting the work into zones and chaining zone to zone, not by SS links with guessed lags. Two zones per floor lets blockwork start in Zone A while the slab in Zone B is still curing; the links are FS and defensible. Zoning also produces the look-ahead the site actually works to.

Floor sizeZonesReason
≤ 800 m²1One pour, one crew, no overlap to model
800–2,000 m²2One slab pour per zone; trades leapfrog
> 2,000 m² or podium3–4Pour size limited by batching and finishing crew

3. Classify every link as you write it

ClassTestLabelExample
HardPhysically impossible otherwiseNone neededStrike formwork after cure; energise after test
Soft — resourceSame crew, set or plant; with a second one the link disappears"SOFT-RES: 2 formwork sets"Formwork L6 waits for strike L4
Soft — preferenceMethod chooses this order; another would work"SOFT-PREF: MS-STR-03 §5.2, top-down finishing"Finishes start from L10 downward
Contractual / accessEmployer or authority dependencyMilestone + Start On or After with document refFaçade waits for podium roof release

Soft-preference is the first lever pulled in compression (schedule-compression); it can only be pulled if it is labelled. A schedule with 3,000 unlabelled FS links has no recovery plan in it.

4. Overlaps that the method statement genuinely contains

Where zoning does not resolve it — a trade that trails another continuously along a façade or a corridor — use an SS/FF pair with a real lag on the predecessor's calendar and a reason (leads-and-lags; dangling-activities). "Ductwork L5 SS+3d Cable tray L5, FF+2d" with Comments "MS-MEP-02: tray leads duct by 3 days, one crew each." Never SS alone (a dangle), never a negative lag, never a lag longer than the predecessor's duration.

5. Cycles — the repeated floor

The typical-floor cycle is the most important sequence on a building. Build it once, prove it, then copy it:

Step (2-zone floor, 8-day cycle)DurationCalendarLink to previousClass
Set out & fix column/wall rebar Zone A16-dayFS strike of the floor belowHard
Column/wall formwork & pour Zone A16-dayFSHard
Slab formwork Zone A (set 1)26-dayFSHard
Slab rebar & MEP embeds Zone A26-dayFSHard
Pre-pour inspection Zone A0.5 → 16-dayFSHard
Pour slab Zone A16-dayFSHard
Cure slab Zone A37-dayFSHard
Strike Zone A16-dayFS cureHard
Zone B chainsameZone B rebar FS Zone A pourSOFT-RES: one rebar crew
Slab formwork L(n+1) Zone A26-dayFS strike L(n−2) Zone ASOFT-RES: 3 formwork sets

The achieved cycle from the last three floors replaces the planned one at every update (forecasting-honestly). The cycle table lives in the SBM with the method statement revision.

6. Hold points, inspections and curing are activities

Inspection requests, third-party tests, authority inspections and curing are the steps most often left out and most often on the critical path. Each is an activity with a duration from the SBM (pre-pour IR 1 day; concrete cube 7/28 days as milestones off the pour; civil defence inspection per the authority's stated period), on the calendar the step actually runs on. Curing is never a lag (calendar-on-relationship-lag).

7. When the method changes

A revised method statement is a logic change: Reflection, links altered, Schedule Comparison, change table row per link, SBM sequence table updated with the new revision number. A method that changed on site without the schedule changing is out-of-sequence progress next month and an unexplained logic edit the month after.

8. Checks before baseline

CheckPass
Every activity traceable to a method statement step or a standard sequenceYes
Every soft link labelledYes (filter on relationship Comments empty where predecessor and successor are different trades)
SS/FF pairs complete; no SS without FFYes
No negative lags; lags ≤ 5% of relationships, each with a reasonYes
Curing, inspections, tests present on the right calendarsYes
Cycle table matches the typical floor logicYes
Method statement revision recorded in SBMYes
🟥 Where people go wrong
  1. Logic drawn before the method is read. Forty FS links on a floor the method says is worked in two overlapping zones; the schedule shows 12 days where the site does 8. Method first, zones second, links third.
  1. Overlaps by negative lag. "Plaster FS −5d Blockwork" instead of two zones. Fails DCMA Metric 2 and hides the real sequence. Zone it.
  1. Nothing labelled. Every link looks hard; compression has nothing to pull; the reviewer can't tell method from physics. Comments field or Logic Basis note on every soft link as it is written — it is thirty seconds each and impossible to do retrospectively on 3,000 links.
  1. Resource logic omitted. Formwork L6 shown starting the day after L5 pours, with two sets. The cycle is impossible and nobody notices until month four. Sets, crews and cranes constrain the sequence; show it.
  1. Inspections and curing missing. The critical path is a day short per floor per omitted step — 14 floors, 3 steps, 42 days. Every hold point is an activity.
  1. Method changes on site, schedule unchanged. Out-of-sequence progress, then a silent re-link. Method revision → Reflection → comparison → change table.
  1. Cycle never re-derived. Planned 7 days, achieved 9 for six floors, schedule still shows 7 for the remaining ten. Achieved rate every update.
⚖️ When you're challenged

"Why can't the façade start until the structure reaches Level 18? That's your logic, not physics." Correct — it's labelled SOFT-PREF, method statement MS-FAC-01 section 4, to keep the crane free for structure. Physics only needs the slab edge two floors above the install level. If we accept the crane conflict, the link moves to L12 and saves twelve days — that's option R1-02 in the recovery table.

"The method statement says three formwork sets. Your programme cycle only works with four." Then one of them changes. With three sets the L(n+1) formwork can't start until L(n−2) strikes, which is a 9-day cycle on the calendar. Either the sub confirms a fourth set in writing and the cycle stays at 8, or the programme goes to 9 and the completion moves. I'm not showing 8 on three sets.

"There's no inspection activity before the pour. The consultant needs 24 hours' notice." Agreed — it's missing on the podium chain. Adding "Pre-pour IR" at one day on each pour, 14 pours, and the podium topping-out moves 14 working days later. Better to find that now than at pour three.

📄 Related pages
✏️ Worked example

Podium slab, mixed-use scheme in Riyadh, 3,200 m² per level, method statement MS-STR-02 Rev 1: four pours per level, one rebar crew, two formwork sets, one concrete pump, 6-day calendar, cure 3 days before back-propped strike.

Element of the methodModelled asClass / label
4 pours (P1–P4)4 activity chains per level, Zone code P1–P4
One rebar crewRebar P2 FS Rebar P1 (and so on)SOFT-RES: 1 rebar crew, MS-STR-02 §3.1
Two formwork setsFormwork P3 FS Strike P1; Formwork P4 FS Strike P2SOFT-RES: 2 sets, §3.2
One pumpPour P2 FS Pour P1SOFT-RES: 1 pump — but trivially satisfied by the rebar link; recorded, not modelled twice (redundant-relationships)
Cure 3 days"Cure Pn" 3 days, 7-day calendarHard
Pre-pour inspection"Pre-pour IR Pn" 1 day, 6-dayHard (contract hold point, ER §4.6)
Cube results7-day and 28-day milestones off each pour, 7-day calendarHard; 28-day result feeds post-tensioning stress activity
Back-propping"Erect back-props Pn" 1 day after strike; "Remove back-props" after 28-day result on level aboveHard, MS-STR-02 §6

Resulting podium level duration: 19 working days per level from first rebar to last strike; three levels overlapped by the set logic → 41 working days for the podium structure. The tender programme had 30, drawn as four parallel pours with no set or crew logic. The 11-day difference was found at baseline, not at pour six.

📖 References
  • FIDIC Conditions of Contract for Construction, 1999 and 2017, Cl 8.3 (check the edition in your contract)
  • NEC3 / NEC4 Engineering and Construction Contract, Cl 31.2 (check the edition in your contract)
  • SCL Delay and Disruption Protocol, 2nd ed. (2017), Core Principle 1; Guidance Part B
  • AACE International RP 24R-03, Developing Activity Logic
  • AACE International RP 38R-06, Documenting the Schedule Basis
  • GAO Schedule Assessment Guide (GAO-16-89G), Best Practice 2
  • DCMA 14-Point Schedule Assessment, Metrics 1–4
  • Oracle Primavera P6 Professional User Guide — Relationships (Comments); Activity Codes; Notebook topics

From the field

Experience from working planners. Unreviewed — read it as experience, not guidance.

Add what you know about construction sequencing and method. What worked, what the consultant pushed back on, what you would do differently next time. A paragraph is plenty.

Contributors get their name and one link on the site — your own templates, course or consultancy. We take nothing and hold nothing.

Add your experience