AI in Facilities
A second failure on the same cooler, lock, or fryer is not another first-time work order. Before the truck rolls, artificial intelligence (AI) should read recent history, flag a new install, and hand the service provider (SP) a packet that aims at the cause.
By Vixxo Facility Solutions
|
90 days Lookback for repeat work on the same asset |
2+ Closed work orders that should change the plan |
30 days Since install, treat it as a possible warranty case |
These are operating rules in a pre-dispatch review, not industry averages.
Repeat failures are rarely about technician skill. More often, the provider never saw the history that would have pointed to the real cause.
Facilities management (FM) directors feel a repeat call as lost customer experience). The store feels it as the same cooler, door, or line down again. The SP arrives without prior notes, without the sibling work order, and without the make or model. A pre-dispatch check pulls parent, child, and recent work on that site and asset. When two or more closed requests share the asset or symptom inside 90 days, the next step is a site call and a wider scope before anyone is assigned. That can mean a camera-and-jet clean when the first visit couldn't see the full line, a repair-versus-replace review, or an install-defect check when the asset is new.
| Check | If it is true | What should change |
|---|---|---|
| 2+ closed requests, same asset, 90 days | This is a repeat failure | Call the site and widen scope before the truck |
| Installed within 30 days | Possible post-install case | Route warranty and the install owner so the right party can address it |
| Missing make, model, serial, access, or contact | The packet is incomplete | Fill the gap before an SP is assigned |
| Open parent or sibling work order | Follow-up may already exist | Link the work. Do not open a duplicate |
Planning weights for why a repeat call stalls. Not a measured field study.
| No recent history attached |
|
45% |
| Dispatched without repeat context |
|
30% |
| Asset identity missing |
|
25% |
Use the pattern as a checklist, then measure your own repeat rate by asset.
Own the outcome on the repair that touches CX. A repeat visit that starts from scratch is a missed opportunity for the whole process, even if the first work order closed cleanly.
The packet splits three jobs: prep before the truck, actions on site, and the asset memory to store after the visit. Fixture-level identity matters. "Men's room toilet 2, with install date and parts warranty" beats "restroom toilet." The SP gets a clearer job, not a thinner one. Providers do their best work when they know what was tried before, so the packet is there to support them, not to second-guess them. The person accountable for dispatch still makes the assignment. AI writes the brief. Same-day quotes still depend on an available technician. Vixxo's network depth, 150K+ technicians and providers, only helps if the right history rides with the work order.
| See how Vixxo uses AI in facilities |
Usually not. Repeat failures most often trace to an information gap in dispatch, such as missing history, a missing asset identity, or an unlinked prior work order. Giving the SP the full picture up front is the fix.
The packet is a brief for review. The dispatcher still assigns the SP. Promote any automated recommendation only after people can see when they overrode it and why.
Same site, same asset or symptom, inside 90 days. A fixture-level name (which toilet, which case, which fryer) keeps a restroom problem from being logged as the whole store.
If the work order opens within 30 days of install, mark it as a possible post-install case. Surface the install owner and warranty status before it is billed as a routine repair.
Fixture identity, what failed, what was done, and any parts warranty date. The next packet should start from that record so the following visit is smarter.
Sources: How Vixxo FM technology works; AI in facilities management; vixxo.com network proof points (150K+ technicians and providers).