Part 2 · 2024 — 2026

Accumulated data becomes process

BIM quantities, once a verification tool on individual sites, became the company's cost management system.

Over the eight years of Part 1, structural frame BIM data generated on every project piled up at Contractor K. Once that accumulation passed a critical point, things that had been impossible became possible.

From here the goal changes. Part 1 asked “are the quantities right?”; Part 2 asks “how do we control cost-ratio risk?” These were the goals Contractor K set, stage by stage.

Goals by stage — what was to be gained from order intake to settlement

BIM goals

BIM-data-driven bid review → support across the whole execution process

Build a transparent, BIM-data-driven quantity management system from order intake to settlement

Order intake

Accurate quantity take-off

· Eliminate risk of omissions
· Prevent over- and under-counting
· Improve cost accuracy
· Secure bid competitiveness

Execution

BIM data integration

· Manage quantity variation
· Make increases visible
· Monitor actual placement
· Data-driven control

Settlement

Transparent settlement

· BIM-based decisions
· Transparent subcontractor settlement
· Evidence for contract increases
· Minimize cost-ratio risk

Accurate forecasting → change management → transparent settlement, to minimize cost-ratio risk

Note that not one item says “let's use BIM well”. Every one is written in the language of cost and risk. BIM appears only as a means, never as the goal.

2.1The workflow changed

The Part 1 process worked on a single site. Today it spans the whole project lifecycle. Same company, same tool — but the point at which it is used is completely different.

Current workflow — one full turn of the frame quantity

↻ Completion data → conceptual estimating DB

Bid · order intake

Frame BIM quantity take-off

General building — BIM modeling

Housing sites — estimated from the accumulated BIM DB

PRE-CON

Preliminary execution quantity

Construction plan reflected and reviewed

Cost risk stripped out before groundbreaking

Frame construction

BIM for final execution

Quantities updated for design changes and site requests

Rebar cost monitoring · continuous

Ordered vs. delivered reconciled

Settlement

Review of settlement documents

Overruns against execution quantity identified

Evidence against over-claimed progress billing

Completion

Captured into the BIM DB

Organized by member type

Becomes the estimating basis for the next project

Each turn makes the next turn more accurate — that cycle is the heart of Part 2.

In the early phase this flow existed only after groundbreaking, and the data stopped when the project ended.

The difference from Part 1 — in the early phase it was used only after groundbreaking, and when the project ended, so did the data. Now it goes in from the bid stage, and completion data flows back into the bid stage.

2.2Who does it — what to keep and what to hand over

You might expect a company with eight years behind it to do everything itself. The opposite is true. Contractor K hands much of the hands-on quantity take-off to specialist service firms. What it keeps is not the work — it is the standards and the judgement.

Who does what by stage — kept in-house vs. handed over

In-house Specialist service firm

Order intake

Review

Groundbreaking

Frame construction

Settlement · completion

In-house

Conceptual estimate · technical proposal

Bid quantity down 2–3%

Model · quantity review

Database & standards

↕ Deliverables reviewed in-house

Specialist service firm

Conceptual estimate from BIM DB

Preliminary execution take-off

Frame BIM modeling

Final execution take-off

Rebar shop drawings

What is held in-house is deciding “what is correct” — the making is handed over.

Frame BIM modeling and quantity take-off are done by specialist firms; in-house sets the standards and verifies the results.

Even eight years in, modeling and quantity take-off are outsourced. What was built in-house is not headcount — it is the standard.

What this means for customers in Japan — you do not need a BIM organization in place before you can start. Contractor K still outsources quantity take-off today. What you need is not internal headcount but take-off results you can trust, and a standard by which to judge them.

2.3Estimating quantities without a model

The most dramatic change came at the order intake stage. Once members from past projects were captured in a database by type, quantities for a new project could be estimated before the detailed drawings even existed.

Past projects captured in a database by member type — transfer columns · unit slabs · isolated footings · core walls · party walls
New sites only need a MASS model for conceptual estimating, mapped to rebar quantities from the member database
Not 1:1 but many:1 matching — accuracy secured by a classification system based on design conditions (floor count, wind exposure, structural scheme)

Better quantity accuracy at the bid stage translates directly into competitiveness, because you no longer have to add contingency you cannot justify.

2.4How close did BIM quantities come to 2D estimating?

Verified against the accumulated data, BIM net quantities reached the following levels versus conventional 2D estimating.

BIM net quantity vs. 2D estimating(2D estimate = 100)

Concrete

97%

3% below the 2D estimate — the customary contingency

Rebar

96%

4% below the 2D estimate

↑ 2D estimate 100%

A lower BIM net quantity means the old estimate contained unjustified slack. Net quantities are taken directly from the model, so the origin of every figure can be traced.

2.5In real time, not at settlement

Conventional cost management happened at points in time — interim and final settlement. By the time a problem showed up, it was often past the point of no return.

The shift in cost management(conceptual)

AS-IS · stage-gate control

Cost ratio

Foundation workInterim settlementCompletion

By the time you see the problem, it is hard to recover

TO-BE · real-time trend control

Target cost

GroundbreakingCompletion

Any drift shows immediately, so the target holds

This is a conceptual diagram of the difference in management approach, not actual data. The point is that the control point moved from settlement to the entire construction period.

What supports this shift in practice is rebar cost monitoring. Begun in Part 1 as a monthly report, it settled into a continuous management system as the data accumulated. What it manages — duplicate ordering across zones, excess contingency rebar, excess temporary fixing bars — is exactly the territory that, in Part 1, “had no standard and so was up for negotiation”.

Crucially, this only works on top of a trustworthy baseline — the BuilderHub quantity. If you cannot trust the quantity you are comparing against, reconciling it more often means nothing.

2.6Results

Results across four sites. These are completion-basis projections, and publicly announced figures.

Savings at completion

approx. 1,300tons

vs. planned quantity

Reduction rate

approx. 5%

four sites combined

In monetary terms

approx. KRW 1.8billion

completion-basis projection

Site	Planned qty (tons)	Delivered per shop drawing (tons)	Reduction
Site A	approx. 5,350	approx. 5,230	approx. 2%
Site B	approx. 10,700	approx. 10,000	approx. 6%
Site C	approx. 4,730	approx. 4,540	approx. 4%
Site D	approx. 6,200	approx. 5,860	approx. 6%
Total	approx. 27,000	approx. 25,600	approx. 5%

A secondary but substantial practical benefit was less time spent analysing settlement causes and preparing paperwork. With the evidence held as data, fewer people were needed to explain the differences.

Part 2 conclusion — conceptual estimates at order intake, real-time control during execution, and transparent evidence at settlement were joined by a single body of data. Accurate forecasting → change management → transparent settlement. None of it could have been built without eight years of data.

The customer and project names are anonymized, and figures are cited only from publicly announced material and within a pre-agreed scope. Diagrams and charts were newly produced from the source data.