Meet Ira · AI Agent

Quality AI Agent

A quality control assistant that works on your inspections, rejections, rework and batch records in DotOne. Ask where rejections are coming from, which vendor's material keeps failing or where a batch went. The agent reads the records, finds the pattern and prepares the hold, rework order or vendor complaint for your quality head to approve.

See the Agent in a Demo Quality Management module

YouWhy have rejections gone up on the pipe line this week?

Reading inspection and rejection entries
Grouping rejections by machine, shift and operator
Tracing rejected batches back to raw material lots
Finding the common factor
Drafting a vendor complaint for approval

Result · Most rejections traced to one PVC Resin lot from a single vendor, with the affected batches on hold and a draft complaint waiting for approval.

Watch the agent work

It finds the pattern. You fix the cause.

Demo data

Agent run · 8:00

Read inspections326 entries
Grouped rejectionsBy machine, shift, vendor
Traced to raw materialLot PR-0912
Held batches3 batches
Drafted vendor complaintWaiting for approval

Rejection rate by machine This week

Extruder 1

4.8%

Extruder 2

1.6%

Moulding 3

2.0%

Printing 1

1.2%

Vendor complaint draftedPVC Resin lot PR-0912 linked to most Extruder 1 rejections.

VendorMetro Chem
Batches affected3

Review ApproveSent to vendor ✓

Batches on hold

B-2291 · PVC Pipe 110 mm · Night shiftHold
B-2294 · PVC Pipe 110 mm · Night shiftHold
B-2297 · PVC Pipe 90 mmRework

The same job, two ways

In a traditional ERP the software stores the data and people do the analysis. In DotOne the agent does the analysis and your team makes the decision.

Traditional ERP

×Collect inspection sheets and rejection registers from each line.

×Type them into a spreadsheet to see totals by machine or shift.

×Dig through GRNs and batch cards to find which material lot was used.

×Hold failed batches by word of mouth and hope nothing is dispatched.

×Write rework instructions and vendor complaints by hand.

With the Quality AI Agent

✓Ask the agent, for example "Where are this week's rejections coming from?"

✓The agent reads inspections, rejections, batches and receipts together.

✓It finds patterns by machine, shift, operator or vendor.

✓It builds the quality report, ready to share or export.

✓It holds failed batches and drafts rework or vendor complaints for your approval.

What it can see

Incoming, in-process and final inspection results
Rejection and rework entries with their reasons
Batches, lots and their production history
Machine, shift and operator for each production entry
Goods receipts and the vendor behind each raw material lot

What you can ask

"Which machine has the highest rejection rate this month?"
"Are rejections worse on the night shift?"
"Which vendor's material fails incoming inspection most often?"
"Which raw material lot went into batch B-2291, and where was it dispatched?"
"Which batches are on hold and why?"

What it can do

Spots rejection patterns by machine, shift, operator and vendor
Puts failed batches on hold so they are not issued or dispatched
Drafts rework orders for approval
Drafts vendor complaints with the inspection results attached, for approval
Answers batch traceability questions from raw material to dispatch

The agent only acts within the roles, permissions and approval rules you set in DotOne. Anything that changes data can be set to need a person's approval.

Frequently asked questions

What does the Quality AI Agent do in DotOne?

It works on your quality records: inspections, rejections, rework and batch history. You ask where problems are coming from, and the agent reads the records, finds the pattern and prepares the next step, such as a hold, a rework order or a vendor complaint.

Can the agent reject material or hold a batch on its own?

Only if you allow it. The agent reads DotOne data and prepares things. It works within the roles, permissions and approval rules you set, and anything that changes data, such as a hold, a rework order or a complaint to a vendor, can be set to need a person's approval.

How does batch traceability work?

DotOne links each batch to the raw material lots, machine, shift and operator used to make it, and to the dispatches it went out in. The agent follows those links, so you can ask where a lot was used or what went into a batch and get the answer in one place.

Do we need to change how we record inspections?

No. The agent reads the inspection and rejection entries your team already makes in DotOne. Recording the reason for each rejection makes its pattern-finding more useful.