Saint John Precision Fabrication Ltd.

Leadership makes sound capital calls on equipment but has no articulated Industry 4.0 or shop-floor-data roadmap. Technology decisions are reactive: a machine is replaced when it fails, a spreadsheet is added when a report is needed. There is no target operating model that says the MES feeds the ERP, the CMMS drives maintenance off runtime, and OEE is the shared operating metric. Ownership of digital initiatives is diffuse, sitting between the plant manager, the controller and an outside IT contractor. The firm has not mapped which data it needs to run the floor in real time, so investments do not compound into a connected system and the OT/IT integration question keeps getting deferred.

Uncoordinated investment means paying twice: for equipment that cannot share data and for the manual workarounds that bridge the gap. Without an OEE target owned at the top, the floor optimises locally and margin leaks stay invisible until the year-end review.

A one-page digital operating model with OEE as the shared metric, a named owner for shop-floor data, and a sequenced two-year plan that connects floor to ERP.

• Low Write a one-page target operating model that defines the MES, ERP, CMMS and OEE data flow and names a single owner for it. 30 days
• Medium Set a plant-wide OEE baseline target and review it in the weekly production meeting so digital investment ties to one number. 60 days

Production counts, downtime reasons and scrap are recorded by hand on travellers and shift sheets, then keyed into Business Central after the fact. There is no MES and no OEE system, so availability, performance and quality are never calculated in real time. Scrap and first-pass yield are known only in monthly aggregate, not by line, product or shift, which is where the actionable variance lives. Machine controllers hold rich runtime data that is never pulled off the PLCs. Reporting is retrospective and manual, so the plant runs on lagging averages. This is the lowest axis at 1.9 and the one that gates every downstream improvement, from planning accuracy to preventive maintenance.

No live OEE means a line can run ten to fifteen points below capability for a full shift before anyone notices. Scrap that is only visible monthly compounds into thousands in wasted material and rework before a root cause is even assigned.

Edge capture off existing PLCs, a real-time OEE board on the two main cells, and scrap and yield reported by line and shift daily.

• Medium Install edge data capture on the two highest-throughput cells and stand up a live OEE board visible on the floor. 90 days
• Low Define scrap and first-pass-yield reporting by line, product and shift, replacing the monthly aggregate view. 45 days

The firm runs a capable ERP for finance, purchasing and inventory, and the equipment base is modern enough to expose runtime data through its controllers. The problem is the middle layer: there is no MES or integration bus connecting the PLCs and HMIs on the floor to the ERP, so data is bridged by people rather than systems. The OT and IT networks are largely flat and unsegmented, which becomes a real risk the moment IIoT sensors or remote vendor monitoring are added. Backups and patching on the office side are reasonable, but the shop-floor devices sit outside any managed lifecycle. The technology is present and mid-pack at 2.8; what is missing is the connective tissue and the network hygiene to make it a system.

The integration gap forces manual re-keying that injects errors into inventory and planning. The flat OT network means one compromised device could reach production control, and a stoppage there is direct lost output measured in hours, not minutes.

A defined MES-to-ERP integration path and a segmented OT network with shop-floor devices under a managed patch and backup regime.

• High Scope an MES or middleware integration path between the PLCs and Business Central, starting with production counts and downtime. 6 months
• Medium Segment the OT network from IT and bring shop-floor devices into a managed patch and backup schedule. 90 days

Operators and supervisors know the product and the machines intimately, and that tacit knowledge carries the plant. What is thin is the analytical layer: no one on staff owns OEE, reads a downtime Pareto, or maintains a CMMS. When the outside IT contractor is unavailable, digital work stalls. Reliance on a few long-tenured people for setup and troubleshooting is a real key-person risk, and little of that knowledge is documented in a way a system could hold. There is appetite to improve, but no defined roles, training path or time carved out for shop-floor-data work. At 2.5 the culture is willing; the capability and the accountability structure to sustain a data-driven floor are not yet in place.

Key-person dependency means a retirement or an absence can slow a whole line. Without an internal owner for OEE and maintenance data, any system that is installed risks going unused, and the investment stalls at go-live.

A named shop-floor-data owner, basic OEE and CMMS literacy across supervisors, and documented setup knowledge that no longer lives in one head.

• Low Assign a shop-floor-data owner and give supervisors basic training on reading OEE and downtime Pareto charts. 60 days
• Medium Document setup and changeover procedures for the top products to reduce key-person dependency. 90 days

This is the strongest axis at 2.9. Work instructions, in-process quality checks and lot discipline are genuinely embedded, and traceability requirements are met, if manually. The weak points are two. First, maintenance is reactive: crews respond to breakdowns and log them informally, with no CMMS scheduling preventive work off runtime hours, so failures that were foreseeable still cause unplanned downtime. Second, quoting, planning and scheduling run on spreadsheets disconnected from real floor capacity and actual job costs, so quotes lean on standard costs that no longer reflect reality and the schedule does not react to a machine going down. The disciplined core is real; the surrounding operational loops are not yet closed or data-driven.

Reactive maintenance turns preventable failures into unplanned downtime, the most expensive hours a plant runs. Quoting off stale standard costs means winning jobs that lose margin and losing jobs that would have earned it, with no feedback loop to correct the estimates.

A CMMS driving preventive maintenance off runtime, and quoting and scheduling connected to real capacity and actual job costs.

• Medium Deploy a CMMS and convert the top failure modes to preventive schedules driven by machine runtime hours. 90 days
• Medium Feed actual job costs back into the quoting model so estimates reflect real floor performance, not standard costs. 120 days

Customers, especially in regulated or export markets, increasingly ask for lot traceability, on-time-delivery data and quality records, and the firm meets these by pulling paper and cross-referencing spreadsheets after the fact. It works but does not scale, and it cannot answer a recall or audit query quickly. There is no systematic capture of on-time-delivery performance, so the firm cannot show or defend its record with a customer. Data governance is informal: no clear ownership of who controls production and quality records, limited access control, and no defined retention or backup policy for the shop-floor data that traceability depends on. At 2.2 this axis reflects a firm that honours its commitments through effort rather than through systems and controls.

A traceability or recall request that takes days of manual searching is a customer-confidence and compliance risk. Without hard on-time-delivery data, the firm negotiates and defends its performance from anecdote, which weakens it in exactly the conversations that set price.

System-held lot traceability and on-time-delivery reporting, with defined ownership, access control and retention for production and quality records.

• High Move lot traceability into the MES or ERP so a full genealogy can be produced in minutes, not days. 6 months
• Low Define data governance for production and quality records: ownership, access control and a retention and backup policy. 45 days