Connect every machine on the floor · including the ones older than the people running them · and turn what they report into numbers your customer’s auditor will accept. Seven of the eleven MES functions are live today, and the first dashboard is on your shop floor in six weeks.
Where is everything, measured?
A laser scan of the line as it stands today, not as the 2011 layout drawing says it stands. Every rung above anchors to this.
Live in Indian plantsSix layers, each consuming the one below it. Nothing on an upper plane is trustworthy if the plane under it is empty. Hover or tap any layer.
Plants across India bought predictive maintenance for assets they could not uniquely name. The dashboards went up. The returns never arrived.
Skipping a rung is the most expensive mistake in industrial digitalisation, because the failure does not surface until month eighteen · after the capex is spent and the programme has a reputation.
In one sentence: we connect the machines you already own, give every one of them an identity, and turn what they report into numbers you can put in front of a customer auditor. No new machines, no rip and replace, and nothing written back to your control system.
Modern PLCs and CNCs connect by configuration. Older machines connect through a driver from our library, a bridge we build, or a retrofit sensor where there is no digital output at all. Machines that cannot be instrumented are still covered through shift entry, so the plant report has no blank rows.
A tag that says PLC4.DB12.INT3 is not data. Before anything is stored, every signal resolves to an asset with a make, a rating, an ISO 14224 class and a place in your hierarchy. That is what makes the number defensible two years later when nobody remembers the tag.
Fixed report packs with the source of every number on the face of the report and every data gap declared instead of interpolated. Production, availability, downtime by reason code, OEE, energy, quality, asset health. Plus an API if your team wants to build their own view.
Wistwin® is a Manufacturing Execution System for Indian brownfield manufacturing. Seven of the eleven MES functions are live in plants today. The remaining four arrive in 2026, and we say which is which rather than leaving you to find out during a supplier audit.
That transparency is deliberate. Most MES programmes in India fail not because the software was wrong but because the plant could not feed it · no asset register, no machine connectivity, no trustworthy data. We start where the failure actually happens.
These four are in build. We are taking a small number of design partners in auto component plants to shape them against real scheduling constraints rather than a textbook. If that is your problem, tell us · you get the roadmap early and it gets built around your line.
Why the order matters: scheduling built on data you cannot trust produces a plan nobody follows. The seven live functions are the foundation the other four stand on, which is why we ship them first and why a plant that starts now is ready the day scheduling lands.
You have an ERP and you are not changing it. Wistwin sits between the shop floor and whatever you already run, so the plant finally speaks to the systems above it in numbers that reconcile.
Asset master synchronised both ways. Production actuals flow back so the ERP sees what the line did, not what the plan said.
If you already run an MES, we feed it clean, asset-resolved machine data instead of competing with it. Many plants keep both.
We read from the history you already own so years of data are not stranded, and write new data to the Wistwin historian.
Your team builds whatever view it wants on the API. No dependency on us for a report we did not anticipate.
This is the whole commercial conversation. The survey puts each machine in a tier, the tier mix sets the price, and you can argue with the mix because the mix has evidence behind it. Nobody quotes you a rate for “legacy equipment” and hopes.
| Tier | What it means | Effort |
|---|---|---|
| T1Connect Direct | Native OPC UA on board. Modern Siemens, Fanuc, Rockwell, newer CNCs. | Configuration only. Days. |
| T2Connect Bridge | Documented protocol, no OPC UA. Modbus, S7, FOCAS, MELSEC, EtherNet/IP. | Driver from our library, or a one-time build that then joins it. |
| T3Connect Adapt | A controller or display exists but offers no open interface. Needs a bridge built for that machine family. | Priced only after a two-day feasibility per family. Never quoted off the survey. |
| T4Connect Sense | No digital output at all. Current clamps, discrete I/O taps, vibration and temperature sensors. | Hardware plus install. Predictable, about an hour per machine. |
| T5Blocked | Connectable but barred · OEM warranty clause, IT policy, safety-instrumented system, machine under live AMC. | No technical work. Needs a written decision from you before commissioning. |
Tier is not a property of the machine, it is a function of how mature our driver library is. The same press is T3 on the first install and T2 on the twelfth. Every bridge we build becomes part of the library at our cost, never billed as custom to whoever asked first.
Sub-minute resolution from T1 to T3 assets. The only level that supports micro-stops, cycle-time variance, live alarms and prediction. Micro-stops are usually where the hidden losses sit.
Per cycle or per event from T4 retrofits. Run and stop, counts, load, energy per unit. Enough for OEE performance and asset health without touching the control panel.
Per shift or per batch through InspectWin™. Availability, counts, reason codes and quality. Under thirty seconds an entry, with a compliance view so your supervisor enforces it, not our support desk.
No instrument on earth tells you a stop was a tool change rather than a material shortage. Reason codes come from the operator at every tier, which is why downtime Pareto works on lines where the machines themselves say nothing.
Every platform built for greenfield assumes a plant full of modern controllers. Indian plants are not that. A typical line has machines from four decades running side by side, and the ones that matter most are usually the oldest. We are built for that mix, and it is the single hardest thing to copy.
Mechanical presses, old lathes, rebuilt imports, machines whose controller died years ago and was replaced with a relay panel. Nothing to read, no port to plug into, and often no drawings left.
The biggest population in Indian plants and the one every integrator underprices. A controller exists and a protocol exists, but there is no OPC UA and sometimes no documentation. Some have a display and nothing else.
Newer Siemens, Fanuc, Rockwell, Mitsubishi and most CNCs bought in the last decade. Already speaking a standard, already exposing structured tags, already ready for exactly this.
Two things follow from that, and both are commercial. You never have to replace a machine to start measuring it. And your plant report has no blank rows on day one · coverage is complete from the first deployment, with resolution varying by what each machine can physically tell us.
Where your data lives is a contractual question in this industry, not a technical preference. A customer audit clause, a PSU policy or a defence requirement decides it, so all three are designed now rather than retrofitted in year two.
Air-gapped plants, defence, PSU, restricted pharma
Everything runs inside your boundary · acquisition, historian, compute and dashboards. No outbound connection required at all. Your IT team holds the keys and nothing leaves the site.
Most multi-plant groups · the common choice
Acquisition, buffering and time-critical compute stay at the edge. Roll-up, cross-plant benchmarking and long-term retention sit in the cloud · ours or your own tenancy. The plant keeps working through a network outage because the edge does not depend on the link.
Single plants, fastest start, lowest entry cost
Managed by us on Indian infrastructure. Nothing to procure, nothing to maintain, fastest path from assessment to a live dashboard. The edge gateway still connects outbound only, so the security review stays short.
The binding rule behind all three: same containers, different orchestration. The on-premise build is never a fork of the product. Two codebases would mean two roadmaps, two support burdens and a product that ages at half speed · so a plant running air-gapped gets the same release as a plant running cloud.
Unbounded scope is what eats the margin on every dashboard project ever run, and it is what leaves you arguing in month four. Everything below is listed in the contract at signature.
Plus role-based views for Plant Head, supervisor and MD, an API for your own analysis, and exports. Anything beyond the contracted packs is a change order at a published rate, stated at signature rather than discovered later.
Not roadmap items. These are how the product behaves on day one, on every rung and in every contract.
We read from your control layer and never write to it, and the gateway connects outbound only · no inbound ports, no VPN into the plant, no firewall exception. Production carries no risk from the install, and the OT security review that usually takes months closes in weeks. That is often what decides whether the project starts at all.
Every data gap is shown on the face of the report instead of quietly filled in. When the network drops for four hours, the dashboard says so and the shift figure stays honest. Most systems interpolate silently, and the first time a supervisor catches it the dashboard is dead. Ours survives that conversation.
Every number carries its source · machine-captured, sensor-derived or human-entered · stamped at capture and printed on the report. When a customer auditor asks where an OEE figure came from, you answer in ninety seconds with a traceable line back to a named asset instead of a week of emails.
A twin holds your layout, asset register, cycle times and downtime history. We publish no customer names, never use your logo to sell to your competitor, and anonymise every reference by cluster and plant type. Hosting is yours to choose · our cloud, your tenancy, or fully on-prem where a customer contract demands it.
You do not buy the ladder and you do not start with the plant. Three entry points, each one self-contained, each one priced off what the step before it found.
VolTwin™ · one clamp, one incomer
A current clamp on the supply gives run, idle, stop, cycle detection and load profile. No panel entry, no OEM warranty exposure, roughly an hour per machine. For a plant with fifty machines nobody can connect, this produces a working dashboard in a week and shows you what the rest of the programme would feel like.
Paid, fixed fee · credited against your order
We walk every machine on the line and tell you exactly what it will take to connect each one, with evidence. You end up with a document you can act on whether or not you proceed with us · and if you do proceed, the fee comes off the order.
AssetTwin™ + ConnecTwin™ · single line or cell
The smallest deployment that produces a trustworthy number. Asset identity first, then live signal resolved to those assets, then one standard report pack. Fixed fee, one purchase order, no board note required. First dashboard on the shop floor by week six.
Why identity is always in the minimum: connectivity without identity produces dashboards nobody can trust. A signal that cannot be resolved to a named, classified asset is a number without an owner · which is exactly how plants ended up with predictive maintenance that never paid back. Pricing is sized to your machine mix and footprint, and the mix comes out of the assessment rather than out of a negotiation.
A Tier-1 asking a Tier-2 for traceability and machine data carries more weight than any statutory notice, because the customer can stop buying. IATF 16949's next revision is expected late 2026 or early 2027 and is understood to cover supply chain traceability, digital connectivity and software quality. The evidence your customer asks for is about to get harder, not easier.
Clusters: Chakan & Pune · Chennai & Hosur · Manesar & NCR · Sanand & Ahmedabad. References by cluster and plant type, never by name.
Each stage is funded on the evidence the last one produced. No big bang, no rip and replace, and no commitment beyond the stage you are in.
Paid, fixed fee, credited against the order if you proceed. We walk every machine and produce a connectivity matrix with evidence, a tag dictionary mapped to asset IDs, and a firm price for the next step. Nothing is quoted before this exists.
Single line or cell, fixed fee. Asset register loaded, signal layer live, one standard report pack running. First live dashboard on the shop floor by week six, acceptance criteria signed at the end.
Every line, every shift, every machine · including the ones with no digital output at all. Edge gateways where the network is intermittent. Plant Head wall, supervisor mobile and MD scorecard live.
Cross-plant roll-up, single sign-on, API, site benchmarking. The point at which head office compares two plants on the same definitions · usually the first thing worth real money above plant level.
These are the questions Indian plant, quality and IT heads actually put to us. Where the honest answer is “no” or “not yet”, it says so.
Start at the rung your plant is standing on, not at the top. Identity precedes signal and signal precedes foresight: a plant cannot predict the failure of an asset it cannot uniquely name. Most Indian plants sit between asset identity and live signal, so the first funded step is usually a unique asset register followed by read-only machine connectivity on one line.
Yes. If it speaks a documented protocol, a driver reads it directly. If it has a display but no open interface, we build a bridge for that machine family. If it has no digital output at all, a current clamp or discrete I/O tap still gives run, stop, cycle count and energy per unit. Nothing is excluded from the plant report · only its resolution changes.
Every data point carries its source, stamped at capture and printed on the report: machine-captured, sensor-derived or human-entered. An OEE figure that is 70% human-entered states that on its face. Data gaps are declared rather than interpolated, so no availability or OEE number is ever published across a network outage without the gap shown. That answers the auditor in ninety seconds.
Typically part-level traceability back to material lot, process records at each manufacturing step, in-process inspection and SPC data, and evidence that manufacturing processes are monitored under IATF 16949 clause 9.1.1.1. The difficulty is rarely the requirement itself · it is producing the evidence from machines that were never instrumented.
No, and by design it cannot. The read-only behaviour is compiled into the driver rather than set by a flag, and the gateway connects outbound only · no inbound ports, no VPN into the plant. Installs happen alongside running production, typically without a shutdown. If you ever want closed-loop write-back, that is a separately scoped engagement with its own sign-off.
The edge gateway buffers to local disk for a minimum of seven days and replays in order on reconnect. Indian plants lose power and lose network routinely, and a system that silently drops data during an outage produces an OEE figure that is quietly wrong. Any gap that cannot be recovered is declared on the report rather than filled in.
There is no fixed threshold, but it must be visible. We show the provenance mix on every report so a 70% human-entered figure is labelled as one. Manual entry cannot produce micro-stops or cycle-time variance at all, so where your business case rests on recovering short stops, those assets need machine capture rather than shift entry.
Not at all · the reality base is an upgrade, not a prerequisite. You can start with an asset register built from a tagging walk and add verified coordinates whenever a capex cycle opens. An asset ID created without a scan accepts coordinates later without rebuilding anything, so nothing you do now has to be redone.
Yes. Seven of the eleven MES functions are live in plants today · data collection, performance analysis, quality, maintenance, product tracking and genealogy, process management and document control. Operations scheduling, dispatch, resource allocation and labour management ship in 2026, and we state that openly rather than leave you to discover it during a supplier audit. If you already run an MES, we feed it clean asset-resolved machine data instead of replacing it.
Verified position and geometry for every asset, which makes route planning, clash detection and remote walkthroughs possible and makes the tagging walk far faster and auditable. It does not by itself produce make, serial, kW rating or ISO 14224 class · those come from the walk. We price scan, walk and connectivity as three clear deliverables rather than one blurred promise.
That is what the architecture is built around. Outbound-only connection means no inbound ports, no VPN into the plant and no firewall exception to justify. Read-only is enforced in the driver. Hosting can be our cloud, your own cloud tenancy, or fully on-prem and air-gapped where a customer contract or a PSU policy demands it.
The 6th Edition took effect on 1 January 2025 and tightened audit enforcement · response to a major nonconformity was cut from 20 days to 15. The next revision, expected late 2026 or early 2027, is understood to update core requirements around supply chain traceability, digital connectivity and software quality. Confirm the final text with your certification body when it publishes.
From FY 2026-27, the top 250 listed entities must obtain assured value-chain data covering at least 75% of purchases from partners representing 2% or more of procurement. If you supply one of them, expect requests for energy, emissions, water, waste and safety figures at plant level, with an assurance trail behind them, even though you carry no SEBI obligation yourself.
An energy baseline gives a working dashboard in about a week. A connectivity assessment takes two to three weeks and ends with a firm price. A single-line deployment puts the first dashboard on the shop floor by week six and signs off at week eight to twelve. Nothing here requires a board note to start.
A score against each of the six rungs, a benchmark against Indian plants in your sector, and the two rungs worth funding next. India work emails only. No sales call until you ask.
India work emails only. No sales call until you ask.
The form did not load. Email connect@groupvb.com or use the contact page.