What I actually do when I'm on your floor.

I stand at the station and time the work at element level — not a shift average, not a standard pulled from a binder written in 2009. Every loss gets a written cause and an estimated time saving, so the opportunity list is ranked by hours before anyone argues about priorities.

When a change is expensive or hard to reverse, I don't ask you to take the estimate on faith. I build the line as a discrete-event simulation, load it with your real timing and variance, and test the change in the model first — so the throughput number is defended before a single fixture moves.

What you get:

• A measured baseline per station, per shift, with variance
• Quota-allocation analysis: what the standard implies versus what the work takes
• An itemised opportunity list with estimated time saving on each line item
• A staffing and capacity model tied to your volume forecast
• A discrete-event simulation of the line to validate a change — and its effect on the constraint — before you commit capital
• Line-balance analysis identifying the true constraint, not the loudest one
• Hypothesis-tested findings, not eyeballed ones: z-tests and design-of-experiments against the data, so a claimed gain survives scrutiny

An improvement that depends on the consultant being in the building isn't an improvement. A management operating system is the set of routines — who meets, when, with what data, and what they're accountable for saying — that holds the gain after handover.

The same discipline runs the projects themselves. I work to a documented project-management process — scoped deliverables, gated stages, and templated close-out — the kind I built and standardised for a capital-projects group, so an engagement lands on a schedule instead of drifting on a retainer.

What you get:

• Structured shift transition agendas, per department, with the specific metrics that shift owns
• Daily huddle formats and a scored assessment sheet so managers can audit huddle quality objectively
• Short-interval control dashboards that report in time to act, not in time to explain
• Current-state and target-state MOS flow maps with a key event schedule to close the gap
• A project-management process and deliverable templates — charters, stage gates, and close-out records — your team keeps using after I leave
• Operational review cadences with a live improvement register, so opportunities are tracked to closure, not to a slide
• Direct field coaching of your supervisors — including the confrontation skills most supervisor training skips

When the answer is hardware, I design the hardware. Drawings to a title-block standard, full bills of material with real part numbers from real suppliers, tolerance and material callouts, and assembly documentation a shop can build from without calling me.

The same engineering scales up to the plant itself. I've produced conceptual facility designs where the process is the building — a wastewater pretreatment system for an automotive plant's effluent, sized and costed — and authored a white paper on reformulating a product line for lower environmental impact. When a process change runs into a permit limit or a sustainability target, that's engineering, not a slogan.

What you get:

• Dimensioned fixture and workholding drawings with section views, revision control, and mass properties
• Indexed bills of material — every fastener, seal, and plate with supplier part numbers and unit pricing
• Material selection with a stated rationale (UHMW, silicone, titanium fastener stock where galvanic or chemical exposure demands it)
• Automation cell conversion layouts, including filtration and containment detail
• Assembly and disassembly process sheets sequenced by operation, with torque specs, fastener sizes, socket callouts, and scrap/reuse flags per part
• Conceptual facilities and environmental process design — pretreatment, effluent, and containment systems sized against the load and the limit

A line is only as controllable as its instrumentation. I work at the control-system layer — the SCADA, the automation controllers, and the protective devices that watch the plant's power and process and act on it in milliseconds — and at the operator layer, where standard work either survives contact with the floor or doesn't.

On the controls side that means real automation-controller work: SEL RTAC logic, IEC 61850 GOOSE messaging between intelligent electronic devices, Modbus register mapping, SCADA screen and tag architecture, and automated protection-system monitoring and verification so a relay misoperation is caught before it trips a line, not after. It's the same instinct as a time study — measure what's actually happening, then make the system act on it — applied to the plant's electrical and control backbone.

What you get:

• SCADA and automation-controller design — SEL RTAC logic, IEC 61850 / GOOSE integration, Modbus register mapping, and tag/screen architecture
• Protection-system monitoring and verification, automated so device health and misoperations surface on their own — informed by event-analysis and synchrophasor (time-synchronised) measurement
• Digital and projected augmented-reality work instruction design, including remote revision workflows that update an entire line in under a day
• Traceability architecture — machine vision and peripheral integration for automated record capture rather than manual logging
• Regulated-change impact analysis, including whether a process modification crosses the threshold requiring a new submission
• Cleanroom-compatible deployment (systems specified against ISO 14644-1 Class 3, US FED STD 209E Class 10, BS 5295 Class 2, GMP EU Grade B)
• Ergonomic and biomechanical workstation assessment — hypothesis-tested against published population strength and dexterity data, not eyeballed