Vertexi5 demo · guided story
From machine data to a maintenance decision
Seven short steps on an oilfield operator's rig, workover hoist and water pump. Every choice you make carries through to the money at the end.
Readings, faults and alerts are simulated in your browser. Every dollar figure is an editable example assumption for discussion, not a measured or expected result. A site survey and the operator's own maintenance records would replace them.
Step 1 of 7 · Equipment
Meet the equipment
Three typical machines on an oilfield site. Pick which go into the pilot. Everything later, including the money, follows this choice.
Try it. Untick a machine and watch the step bar: the number of assets and the illustrative yearly value change.
Step 2 of 7 · Data ports
Find the data that already exists
Most of what we need is already inside the machines. Pick a point to see where its data comes from and how we would plug in. Amber points have no sensor yet.
| # | Source | Language | Signals | Status |
|---|
Typical for this class of equipment. A one-day site survey would confirm the exact ports, models and access before anything is installed. Sensor prices are example assumptions.
Step 3 of 7 · Connect
Connect it, safely
A small rugged computer on site, the edge box, collects the data and sends it over 4G or satellite. Connect each machine, then cut the network to see what happens.
How a reading travels
How a pilot would be set up
Design intent for a pilot, confirmed in the site survey. Not a certification.
- Read-only. We listen to the machines. Nothing installed can start, stop or change them.
- Keeps working offline. The edge box holds readings while the link is down and sends them when it returns.
- Encrypted in transit. Data travels over TLS to a cloud region or a server the operator chooses.
- Separate network. The edge box sits on its own network segment.
Step 4 of 7 · Live data
Watch live data react
Change how a machine is working and watch its simulated readings respond within seconds. This is the view a team would have on site and on a phone.
Step 5 of 7 · Spot it early
Spot a problem days before it happens
Pick a machine and let a simulated fault develop. Each click moves time forward a day. The alert fires when the trend leaves the learned normal band.
Step 6 of 7 · Decide
Decide what to do
With days of warning, the team chooses when to stop the machine. Compare a fix at a planned stop with running until it breaks. Costs are example assumptions you can edit in step 7.
Step 7 of 7 · Cost vs benefit
What it could cost and what it could return
Pilot cost follows your choices in steps 1 to 3. Benefit comes from failures caught early. Every figure is an illustrative, editable example assumption, not a result.