United States Air Force – BUILDER Data Confidence Dashboard

BUILDER Data Confidence Dashboard Pinpoint which AIR FORCE Facilities Have Good Data

SUMMARY:  The United States Air Force is working to mature its Asset Management processes, procedures, and strategies to an end-state that optimizes life-cycle management while balancing risk and fiscal constraints.  The Sustainment Management System (SMS) BUILDER experts came together over the course of a year and a half to transform how we measure confidence in SMS data transforming building-level decisions.

DESCRIPTION:  Limited infrastructure funding demands smarter facility sustainment and repair decisions to keep power projection platforms performing at capacities required by today’s evolving missions. Smart decisions start with accurate data.

By adapting industry concepts to Air Force needs, the team devised a revolutionary metric combining 13 measures of data confidence, transforming lengthy lists of anomalies from data quality reports into a straightforward rating for each facility. This metric delivers a simple guide to which building condition data can be trusted and which is suspect.

BUILDER data is used for decisions daily at all levels of the enterprise. The Data Confidence metric bridges the gap between collecting data and making decisions by summarizing the enterprise’s trust in the data beforehand. Analysis demonstrated to Higher Headquarters how BUILDER data is good enough for enterprise analysis despite issues with many individual buildings and how targeted initiatives are already improving confidence numbers.

The team developed a user-friendly dashboard to automate the calculation of the Data Confidence metric and allow decision makers at all levels to interact with their BUILDER data and investigate potential quality issues. It gifted decision makers with a BUILDER expert’s intuition about data’s suitability for decisions. Data Confidence dashboard connects with 3 other Facilities Activity Management dashboards for full spectrum visibility of vertical assets.

In the Air Force VAULT, decision makers can get an instant, automated read on the data confidence for any of 35,000 buildings they need to decide on. This previously could only be done manually by a handful of Air Force BUILDER experts.


Anyone who has interacted with BUILDER Sustainment Management System data has sensed the obvious potential. The Army-developed, OSD-mandated database’s ability to track the condition of every facility in the Air Force down to individual components, to project how that condition will degrade over time, and to predict the costs needed to sustain and repair the facility, are powerful capabilities for asset management. Yet anyone who has interacted with BUILDER has also quickly realized this power is only effective if the data used to populate BUILDER is accurate. Garbage in, garbage out, as they say. 

The fact is, BUILDER data is being used by decision-makers at all levels of the Civil Engineer enterprise on a daily basis. Even if some of the data is suspect, we should not throw out all of it. So how do we identify where we can trust the data and where we should exercise caution? That was the problem statement for the BUILDER Data Confidence Working Group, a team of Air Force Installation and Mission Support Center and AFCEC experts on BUILDER and data science, who developed the newly released BUILDER Data Confidence dashboard.

After all, expert BUILDER analysts can tackle this exact problem on a case-by-case basis. Present them with a building’s data, and they can point to anomalies in the data that reduce their confidence in the accuracy of the Building Condition Index, or BCI, outputted by BUILDER. The Working Group adapted industry concepts to automate this expert skill for decision-makers t all levels and summarize these anomaly lists with a simple confidence rating for a building’s BCI of High Confidence, Medium Confidence, Low Confidence, or No Confidence.

This information is now available in a user-friendly dashboard on the Air Force VAULT for anyone with VAULT access. (The link is https://tableau.afdatalab.af.mil/#/site/AFIMSC/workbooks/1807/views.) There, users can filter the Overview page’s list of 35,000 Air Force vertical facilities down to the installation, Category Code, or Mission Dependency Index, or MDI, tier they are interested in. Once they have honed in on a particular facility, they can jump over to the Drill Down page to see a detailed breakdown of the particular anomalies driving the calculated confidence level. Those anomalies fall into 13 types, as identified by Air Force BUILDER experts, and each type receives a confidence rating. The overall confidence in the BCI is then the worst offender of these 13 anomaly types.

No data entry occurs in the dashboard, and even those without BUILDER access can use it to dive into the intricacies of the authoritative database. That said, if a user’s investigation uncovers data issues that need to be fixed they must go into BUILDER itself to make the changes. The Data Confidence dashboard is connected to three other Facilities Activity Management dashboards for full spectrum visibility of vertical assets.

Ultimately, the BUILDER Data Confidence dashboard is a guide, not a verdict on the data. It does not have all of the context a decision-maker uses, only what is contained within BUILDER. In addition, the confidence rating only refers to the likely accuracy of the BCI itself, not the confidence rating in a particular decision, such as whether to demolish, repair, or replace. Still, it can provide reassurance that the starting point for many decisions, the condition of the facility, is reliable, resulting in more confident decision-making. 

If you would like to learn more about the BUILDER Data Confidence dashboard, please reach out to Ben Graf (ben.graf@us.af.mil) and John Glass (john.glass.1@us.af.mil) at the Air Force Civil Engineer Center (AFCEC/CPR). 

Photo of BUILDER Data Confidence Dashboard

Scroll to Top