ATLAS / BLOG
FROM THE
WORK.
Notes on coordination, construction technology
and the decisions behind the work.
Overloaded-team coordination
What to hand off first when coordination is compressing
The release date moved and the coordination lead is already committed. Choose a first assignment that outside help can start without draining the person it is meant to relieve.
6 MIN READStaff augmentation versus outsourcing
Embedded help and an outsourced package solve different problems
Staff augmentation adds hours inside your workflow. An outsourced package buys a defined output. How to tell which one the work needs, and what to agree before anyone opens a model.
6 MIN READTechnology adoption mid-project
Introducing a new tool without stalling the work
A bounded trial lets a project team test new software or a new workflow on real work, with a recorded baseline, a way back and a decision made on evidence.
6 MIN READArtificial intelligence with restricted data
Running artificial intelligence on a job with restricted data
Before drawings, specifications or correspondence go into an artificial intelligence tool, map where the information travels, who sets the rules and how the output will be checked.
6 MIN READWhat research supports about rework hours
How many hours does construction lose to rework and information friction
The 14.1-hour figure is a 2018 survey composite, not a measured rework rate. Here is what surveys and recorded studies each show, and how to test a statistic before it goes into a business case.
8 MIN READMeasuring your own team's effort
How to measure your team's coordination effort and test an improvement
A proposed Atlas method for a local baseline: agreed activity categories, one record per correction, labor kept apart from waiting, and changes judged by accepted output.
6 MIN READ