What is manufacturing operational intelligence?

Short answer. Manufacturing operational intelligence is software that continuously reads a plant's operational records, connects them across documents and time, and tells people which problems are about to cost production time or a customer delivery — with the reason, the evidence and how sure it is. It supports decisions. It does not replace the ERP that holds the records, the MES that runs the shop floor or the planner who decides what to change.

Where the term comes from

“Operational intelligence” describes analytics that work on live operational data rather than on last month's reports. In large manufacturing, the term is often used for platforms built on machine and historian data next to an MES.

Many small and mid-sized manufacturers have no MES and no historian. What they do have is an ERP — often ERPNext — holding sales orders, work orders, job cards, BOMs, stock and purchase orders. For them, operational intelligence starts from those transactions: the same question, “what is about to go wrong?”, answered from the records the plant already keeps.

What it does

Five capabilities separate operational intelligence from a report or a chart:

1. Connects the chain. Customer order → work order → operation → material → supply, including internal supply from other work orders.
2. Reasons in time. Compares when something is needed with when it can credibly be ready, using calendars and planned dates.
3. Ranks by consequence. Puts the problem that threatens the earliest or most important customer promise first, instead of listing every exception.
4. Explains itself. Every finding shows the documents and quantities behind it, so a person can check it in minutes.
5. Checks its own inputs. Says when the data is stale, incomplete or inconsistent, and lowers its confidence accordingly. See Stale ERP data creates false confidence.

How it differs from systems you may already have

System	Its job	What it does not do
ERP (ERPNext)	System of record: orders, BOMs, stock, accounting	Join records into consequences across documents and time
MES	Run and record shop-floor execution: dispatching, data capture, traceability	Usually works from machine and operator data rather than ERP order chains, and is typically a larger project to deploy
APS / scheduler	Create or optimise a production schedule	Explain today's risks in the current plan; it changes the plan
BI dashboards (e.g. Frappe Insights)	Chart the queries you design	Decide the joins, timing rules and priority for you
Supply-chain control tower	Cross-company visibility over suppliers, logistics and inventory	Usually enterprise-scale, multi-system, rarely aimed at a single plant on ERPNext
Operational intelligence	Find and rank the problems about to cost time or deliveries, with evidence and confidence	Execute anything: people decide and act in the ERP

These are complementary. A plant can use ERPNext as its record, a simple dashboard for status, and an operational intelligence layer for consequence.

What it is not

• Not magic prediction. Good operational intelligence can be entirely rule-based. Deterministic rules are easier to check and trust than an opaque score.
• Not automation. It points at what to check first. It does not reschedule, reorder or change documents.
• Not a replacement for knowing the floor. ERP data is often behind reality; a good tool says how far behind it might be.

On top of ERPNext

ERPNext already holds what this needs: Sales Orders with delivery dates, Work Orders and their operations, Job Cards, BOMs, stock per warehouse, Purchase Orders and Purchase Receipts. With a dedicated read-only API user, an external tool can read them without installing anything inside ERPNext and without any ability to change them. The difference between what ERPNext's own dashboards show and what a consequence view adds is explained in ERPNext manufacturing visibility: status is not consequence.

Where TaktSignal fits

TaktSignal is Manufacturing Operational Intelligence for ERPNext. It reads sales orders, work orders, BOMs, stock and purchase orders with a read-only account, runs on your own computer, and turns them into ranked operational risks — material shortages, overloaded work centers, late or blocked work, customer orders at risk — each with its evidence, the time at stake and a confidence level. The rules are deterministic and documented; no AI is required. Understanding alerts.

It is a Community Alpha: tested with synthetic factories and against ERPNext 15 and 16, not yet validated in real factories.

FAQ

Is manufacturing operational intelligence the same as a control tower? They overlap. “Control tower” usually means enterprise supply-chain visibility across companies and systems. Operational intelligence for a single plant focuses on that plant's orders, production and material.

Do I need an MES first? No. When the ERP already holds work orders, BOMs, stock and purchasing, there is enough to find most material and delivery risks. Shop-floor data improves timing; it is not a prerequisite.

Does it need AI? No. Explainable rules over ERP records answer most “what is about to go wrong?” questions, and they make every result checkable.