Manufacturing performance, end to end.
Stone Consulting brings senior operating experience to food, beverage, and discrete manufacturers. We walk the plant, model the constraint, fix the data, rewire the process, and apply AI where it pays back. Then we stay until the numbers move.
How we help clients
Most firms send a specialist for each problem. We bring one operator who has run procurement, the shop floor, the warehouse, the network, and the systems behind them, so the fix in one area doesn't break another.
Supply chain, procurement, and contract reviews
Sourcing and contract re-bids, supply and demand modeling, MRP optimization, cycle counting, and inventory reduction.
ERP, WMS, MES, and Power BI
Full-cycle system implementations and live dashboards that replace the monthly report.
AI where it pays back
Predictive maintenance, computer-vision quality, gen-AI work instructions, and ML forecasting, sequenced after the data and process are fixed.
OEE, downtime, yield, and plant diagnostics
Plant walks, loss trees in dollars, daily management, quality and sanitation, OSHA-aligned safety, and PM/TPM programs.
Footprint, freight, and distribution
Plant and hub strategy, carrier contracts, cold chain, DSD route design, and spare parts distribution.
Bottleneck analysis, layouts, and CI
Constraint analysis, CAD line layouts, labor standards, takt-based line design, floor and maintenance assessments, and multi-plant CI systems.
Walk. Model. Fix. Sustain.
The same four moves on every engagement, from a single line to a multi-plant network. No phase ends without a number attached to it.
Walk
One to two weeks on the floor. Rate checks at every station, time studies, data audit, and a loss tree in dollars.
Weeks 1–2Model
Constraint model, layout, and dashboard spec. Every option ranked by throughput bought per dollar spent.
Weeks 2–4Fix
Ninety-day sprint on the top three losses with named owners, a weekly number, and a live screen at the tier meeting.
Days 30–120Sustain
Standards, PM discipline, and a CI system your people run. We coach until the gains hold without us.
Days 120–180Results typically include
- 8–25 pts OEE on the constraint line
- 10%+ savings on sourced categories
- 20–40% reduction in unplanned downtime
- 99%+ inventory accuracy
- 10–20% lower cost per case delivered
- 90 days to a live dashboard the plant runs on
Ranges reflect outcomes from Brandon's operating career and typical engagement scope. Every plant is different; the diagnostic sets your targets.
AI that works on a real plant floor.
Most AI pilots stall on bad data and broken processes. We sequence it the other way: fix the data, fix the process, then apply AI to the losses that are left. That's how it pays back in months, not years.
Predictive maintenance
Sensor and PLC data feeding failure prediction on the assets that drive your downtime Pareto.
Computer-vision quality
Camera-based inspection for fill levels, seals, labels, welds, and finish, with closed-loop containment.
Gen-AI on the floor
SOPs, work instructions, and troubleshooting guides operators can ask questions of, in their language.
ML forecasting and inventory
Demand forecasting and safety-stock optimization that feed MRP directly instead of a spreadsheet.
Digital twin and scenarios
Line, plant, and network models to test layouts, capacity, and footprint decisions before spending capex.
Unified namespace and IT/OT
One data architecture from PLC to ERP so every dashboard and model uses the same numbers.
One dashboard for every discipline.
Brandon has spent over 10 years building Power BI dashboards for operations, supply chain, systems, distribution, engineering, and the executive P&L, all on one data model, refreshed in real time. Every engagement leaves one behind.
Plant Performance
Week 36 · Refreshed 6:02 AMOEE by line vs. target (85%)
OEE trend, 12 weeks
Downtime Pareto (hrs, this week)
| Line | Availability | Performance | Quality | Scrap % | PM compliance | Status |
|---|---|---|---|---|---|---|
| Line 1 | 94.0% | 96.1% | 97.4% | 1.2% | 100% | On plan |
| Line 3 | 87.5% | 90.2% | 96.3% | 2.4% | 92% | Watch |
| Line 5 | 78.1% | 85.0% | 96.4% | 3.9% | 71% | Action |
Supply Chain Control Tower
Week 36 · Refreshed 5:45 AMInventory health by class (% within target)
Forecast vs. actual, 12 weeks (K cases)
Expedites by cause (this month)
| Supplier | Category | Spend (YTD) | OTD | Quality PPM | Contract expiry | Status |
|---|---|---|---|---|---|---|
| Supplier A | Packaging | $24.1M | 98% | 120 | Mar 2027 | On plan |
| Supplier B | Sweeteners | $18.7M | 89% | 340 | Nov 2026 | Re-bid |
| Supplier C | Films | $9.3M | 81% | 610 | Oct 2026 | Action |
Systems & Data Health
Week 36 · Refreshed 2 min agoIntegration health by interface (% success)
Dashboard active users, 12 weeks
Master data defects by type
| System | Go-live | Users | Uptime | Open tickets | Owner | Status |
|---|---|---|---|---|---|---|
| ERP | Phase 2 live | 412 | 99.9% | 7 | IT + Ops | Stable |
| WMS | Live | 118 | 99.7% | 3 | Warehouse | Stable |
| MES | Line 5 pilot | 36 | 98.2% | 12 | Engineering | Pilot |
Distribution & Freight
Week 36 · Refreshed 5:30 AMCost per case by hub (vs. $1.50 target)
Freight spend vs. budget, 12 weeks ($K)
Late deliveries by cause (this month)
| Lane / route | Mode | Cases (wk) | Cost / case | OTIF | Utilization | Status |
|---|---|---|---|---|---|---|
| Hub → Retail DSD (fresh) | DSD | 48,200 | $1.31 | 97% | 91% | On plan |
| Hub → Retail DSD (frozen) | DSD | 31,600 | $1.58 | 95% | 84% | Watch |
| Plant → Hub transfers | LTL | 62,900 | $0.44 | 98% | 93% | On plan |
Engineering & Continuous Improvement
Week 36 · Refreshed 6:10 AMCI savings by plant (% of annual target)
Cumulative CI savings, 12 months ($K)
Projects by type (active)
| Project | Plant | Type | Owner | Savings (annualized) | % complete | Status |
|---|---|---|---|---|---|---|
| Filler changeover SMED | Plant 1 | SMED | J. Rivera | $420K | 80% | On track |
| Kitting cell redesign | Plant 3 | Layout | M. Chen | $310K | 45% | Watch |
| Line 5 labor rebalance | Plant 5 | Labor | A. Patel | $275K | 20% | Behind |
Executive Scorecard
Month: August · Refreshed 6:00 AMConversion cost per case by plant (vs. $4.00 target)
EBITDA vs. plan, 12 months ($M)
EBITDA bridge, YTD ($M vs LY)
| Plant | Revenue | Conv. cost / case | OEE | Labor % | EBITDA margin | Status |
|---|---|---|---|---|---|---|
| Plant 1 | $41.2M | $3.56 | 88% | 14.1% | 16.8% | Ahead |
| Plant 4 | $27.8M | $4.16 | 76% | 17.9% | 10.2% | Watch |
| Plant 5 | $19.4M | $4.72 | 64% | 21.3% | 4.1% | Turnaround |
People & HR Analytics
Month: August · Refreshed 6:15 AMTurnover by plant vs. target (15%)
Headcount vs. plan, 12 months
Terminations by reason (YTD)
| Plant | Headcount | Hires (YTD) | Terms (YTD) | Turnover | Overtime % | Absenteeism | Status |
|---|---|---|---|---|---|---|---|
| Plant 1 | 612 | 98 | 74 | 12.1% | 6.2% | 2.8% | Healthy |
| Plant 4 | 388 | 92 | 78 | 19.4% | 11.8% | 4.9% | Watch |
| Plant 5 | 296 | 134 | 118 | 31.2% | 16.4% | 7.1% | Action |
Sanitation & Food Safety
Week 36 · Refreshed 5:50 AMPre-op pass rate by line vs. target (98%)
ATP swab pass rate, 12 weeks
Pre-op failures by cause (this month)
| Line | Pre-op pass | ATP pass | Sanitation hrs | Clean-to-run (min) | Master schedule | Status |
|---|---|---|---|---|---|---|
| Line 1 | 100% | 98% | 62 | 48 | 100% | On plan |
| Line 4 | 96% | 94% | 74 | 66 | 92% | Watch |
| Line 5 | 91% | 89% | 88 | 84 | 78% | Action |
Service Level Summary
Week 36 · Refreshed 6:05 AMCase fill rate by customer vs. target (98%)
Lost sales, 12 weeks ($K)
Cuts by root cause (cases, K)
| Customer | Ordered | Shipped | Cuts | Add-outs | Lost sales | Fill rate | Status |
|---|---|---|---|---|---|---|---|
| Customer A | 48,200 | 47,720 | 480 | 110 | $14K | 99.0% | On plan |
| Customer C | 31,600 | 30,650 | 950 | 390 | $41K | 97.0% | Watch |
| DSD retail | 62,900 | 59,130 | 3,770 | 1,420 | $131K | 94.0% | Action |
Illustrative dashboards with sample data. Switch tabs to see one view per discipline, plus executive, HR, sanitation, and service-level views. Client dashboards are built on live plant, ERP, WMS, and MES data.
Ambition in action
Selected results across food, beverage, and discrete manufacturing. Client names withheld.
Sourcing reset across raw materials
Consolidated category spend across plants, re-bid every major contract, and put price, volume, and expiry on a live dashboard.
Turnaround ahead of a PE exit
Rebuilt daily management around OEE, yield, and labor per case; restored PM compliance; returned the plant to plan.
Engagement models
Sized to the problem. Fixed scope, fixed fee where possible, and a results-based component when the numbers are measurable.
Plant diagnostic
Plant walk, rate checks, loss tree in dollars, and a ranked plan. The fastest way to know where the money is.
Performance sprint
Top three losses, named owners, weekly number, live dashboard. Fee partly tied to the result.
Interim or fractional leadership
Plant manager, director of operations, or supply chain lead while you hire, or through a turnaround.
Diligence and 100-day plan
Operational diligence, value-creation case, and the 100-day plan, written by someone who has been through multiple exits on the operating side.
Systems and dashboards
ERP, WMS, or MES implementation ownership, and Power BI builds on one data model.
Retained advisor
Monthly on-site day plus on-call access for the leadership team and the board.
What you get that a big firm can't offer.
Large firms staff manufacturing work with rotating teams and a partner who visits monthly. We do it the other way around.
What leaders say after the numbers move.
From plant managers, finance leaders, and sponsors we have worked with.
"Brandon found our real bottleneck in the first week. It wasn't the one we'd been chasing for two years. Ninety days later the line was running 18% more cases with the same crew."
"He walked the floor with our team instead of sending analysts. The dashboards he built are still what we run the plant on, and he didn't leave until the numbers held for three months."
"The diligence read like it was written by the person who'd have to run the plant, because it was. We used his 100-day plan as the value-creation case and it held through exit."
"Our raw material contracts had been rolling over on autopilot. He re-bid the top categories and took over 10% out without a single supply disruption."
"We had sixteen ways of measuring efficiency across the network. He gave us one, and got the plants to actually adopt it."
"He balanced our welding, powder coat, and assembly to one takt time. Lead times on custom units went from unpredictable to something we could promise a customer."
Featured insights
Points of view from the plant floor on AI, planning, and performance.
Why most manufacturing AI pilots die in month four
The model was fine. The downtime codes were not. What to fix before you buy anything.
Your MRP isn't broken, its parameters are
Lead times, safety stocks, and lot sizes nobody has reviewed in years are why planners override the system.
The one number every plant should manage
OEE is a diagnostic, not a target. How to pick the metric that actually moves the P&L.
Connect with our practice leader
One senior operator, on site, accountable for the result.
Brandon Stone
Start with a plant walk.
One or two weeks on site, a data pull, and a ranked list of what's costing you the most. Then we agree on the one number we're going to move.
Five disciplines. One accountable operator.
Every area below is work Brandon has done himself as an operator holding the P&L, not observed from the outside. Open each discipline to see the services and what good looks like.
SSupply chain
Get the right material in the door at the right cost, and keep inventory honest.+
Supply chain
Procurement and sourcing
- Category strategy and raw material procurement (nine-figure spend managed)
- Contract negotiation and supplier consolidation
- Supplier scorecards, quality agreements, and risk management
- Packaging and ingredient spec standardization
- Tariff and landed-cost modeling
Planning and inventory
- Full supply and demand models, forecast to production schedule
- S&OP / IBP design and facilitation
- MRP strategy: parameters, lot sizing, lead times, safety stock
- Cycle counting programs and 99%+ inventory accuracy
- ABC classification and obsolete inventory reduction
What good looks like
- One demand number shared by sales, ops, and finance
- Planners trust MRP and stop overriding it
- Cycle counts replace the annual physical
- Contract expiry and PPV visible on a live dashboard
TTechnology
Systems that go in without a shutdown, and dashboards people actually use.+
Technology
Systems implementation
- ERP selection and full-cycle implementation
- WMS selection and go-live across cold and dry storage
- MES and shop-floor data capture
- Master data cleanup: items, BOMs, routings
- Integration across ERP, WMS, MES, and planning tools
Analytics and visibility
- Power BI development, 10+ years of live dashboards
- Plant, supply chain, and P&L performance models
- Real-time OEE, downtime, and labor reporting
- Data architecture, unified namespace, refresh automation
- Executive scorecards and board reporting
What good looks like
- Go-live with zero lost production days
- One source of truth from PLC to P&L
- Tier meetings run off a live screen, not a printout
- Your team owns the data model when we leave
OOperations
Run the plant on numbers, every shift, with the people and lines you already have.+
Operations
Performance and cost
- Plant turnarounds and cost-out programs
- OEE: availability, performance, and quality losses
- Downtime tracking, root cause, and Pareto-driven action
- Yield, waste, and shrink reduction
- Labor standards, crewing, cases per labor hour
- Changeover reduction (SMED), schedule attainment
- Takt-based assembly, sub-assembly, and kitting
- Cold, dry, and warehouse operations
Quality, safety, maintenance
- Quality systems: HACCP, GMP, SQF for food; ISO for discrete
- Hold, rework, and complaint reduction
- OSHA-aligned safety programs, TRIR and DART reduction
- Preventive maintenance program design and compliance
- TPM, autonomous maintenance, operator care
- MTBF, MTTR, and spare parts planning
- Daily management, tiered meetings, visual boards
What good looks like
- Every line has a target, a live number, and an owner
- Downtime Pareto drives the week's maintenance work
- PM compliance above 95%, reactive work falling
- Safety incidents anticipated, not investigated
NNetworks
Design the footprint and the freight behind it.+
Networks
Footprint and distribution
- Manufacturing footprint and plant consolidation
- Distribution hub and cross-dock strategy
- Cold chain design for frozen and fresh
- Make-vs-buy and co-packer strategy
- Network modeling and scenario analysis
Freight and last mile
- Freight strategy, carrier bids, multi-year contracts
- DSD route design and fleet utilization
- Spare parts and aftermarket distribution
- Cost per case delivered and OTIF improvement
What good looks like
- Footprint decisions tested in a model before capex
- Freight under contract, re-bid on a cadence
- Fresh and frozen routed separately and full
- Service orders no longer pull from production
EEngineering
Industrial engineering and continuous improvement that make gains stick across every plant.+
Engineering
Industrial engineering
- Labor standards, time studies, line balancing
- Plant layout and material-flow design
- Takt time and cell design for assembly lines
- Equipment specification, capex justification, commissioning
- Fabrication, welding, powder coating, upholstery process design
Continuous improvement
- CI systems across multi-plant networks
- Lean, Six Sigma, and kaizen leadership
- Value-stream mapping and standard work
- Plant scorecards and CI project tracking
- CI lead development and coaching
What good looks like
- One set of standards across every plant
- Improvements travel between sites within a quarter
- Capex justified on measured losses, not estimates
- Plant CI leads run the system without us
Everything a manufacturer asks us for.
Filter the list, or scan it. If it happens inside a plant, a warehouse, or a network, it's here.
Assess & diagnose
- Plant diagnostic and loss tree
- Bottleneck and constraint analysis
- Capacity and throughput modeling
- Production floor assessment
- Maintenance assessment
- Supply chain review
- Procurement and contract review
- Network and freight review
- Systems and data assessment
- Maturity assessment (STONE)
- PE operational due diligence
- Quality, sanitation, and safety audit
Supply chain & procurement
- Category strategy and sourcing
- Contract negotiation and re-bids
- Should-cost modeling
- Supplier consolidation and scorecards
- Supply and demand modeling
- S&OP / IBP design
- MRP parameter optimization
- Inventory reduction and ABC policy
- Cycle counting programs
- Packaging and ingredient spec standardization
- Tariff and landed-cost modeling
- Supplier risk management
Operations & performance
- OEE improvement
- Downtime reduction and root cause
- Yield, waste, and shrink reduction
- Changeover reduction (SMED)
- Labor standards and crewing
- Cases per labor hour
- Schedule attainment
- Daily management and tier meetings
- Visual management and 5S
- Lean transformation and kaizen
- Six Sigma problem solving
- Plant turnarounds and cost-out
Quality, safety & sanitation
- HACCP, GMP, and SQF programs
- ISO-aligned quality systems
- Hold, rework, and complaint reduction
- Sanitation and SSOP programs
- Pre-op and ATP verification
- OSHA-aligned safety programs
- Near-miss and leading indicators
- Allergen and changeover controls
Maintenance & reliability
- Asset criticality ranking
- PM program design and compliance
- TPM and operator care
- Planning and scheduling
- Spare parts strategy
- MTBF / MTTR improvement
- Condition-based monitoring
- Predictive maintenance
Engineering & layout
- Line balancing and takt design
- Plant layout (current and proposed, CAD)
- Material flow and WIP design
- Time studies and standards
- Equipment specification and commissioning
- Capex justification and ranking
- Fabrication, welding, and finishing process design
- Digital twin and simulation
Technology & data
- ERP selection and implementation
- WMS selection and go-live
- MES and shop-floor data capture
- Master data cleanup and governance
- Systems integration and unified namespace
- Power BI dashboard development
- Executive scorecards and board reporting
- AI use-case selection and deployment
Networks & distribution
- Manufacturing footprint strategy
- Distribution hub and cross-dock design
- Freight strategy and carrier RFPs
- DSD route design
- Cold chain design
- Cost per case delivered
- OTIF and service-level improvement
- Spare parts and aftermarket distribution
People & leadership
- Interim and fractional plant leadership
- 100-day plans
- HR analytics and turnover reduction
- Crewing and shift design
- Training and standard work
- CI lead development
- Leadership tier coaching
- Change management on the floor
What we measure
The metrics every engagement is built around. If it isn't on a dashboard by week two, it isn't being managed.
Have a specific problem in mind?
AI that pays back on the plant floor.
Manufacturers don't get value from AI by dropping models into broken operations. We sequence it: fix the data, fix the process, then point AI at the losses that remain. Every use case below is tied to a line on the P&L.
Our sequence: data, process, then AI
The order matters. Skipping a step is why pilots stall.
Fix the data
Downtime reason codes, master data, item and BOM accuracy, PLC tags, and a unified namespace so every system reports the same number.
Fix the process
Daily management, standard work, and PM discipline. AI can't predict a failure on a machine nobody maintains.
Apply AI to what's left
Point models at the top of the Pareto: the asset, the defect, the SKU, or the route that still costs the most.
AI and digital use cases
Chosen for payback, deployed with the operators who will use them.
Predictive maintenance
Vibration, temperature, current, and PLC signals feeding failure prediction on the assets at the top of your downtime Pareto. Integrated with the PM schedule and CMMS.
- Reduces unplanned downtime and reactive work
- Extends MTBF on critical assets
Computer-vision quality
Camera inspection for fill level, cap and seal, label placement, weld quality, and finish defects, with closed-loop containment so bad product stops at the station.
- Cuts holds, rework, and customer complaints
- Replaces sampling with 100% inspection
Gen-AI operator assistant
SOPs, work instructions, changeover guides, and troubleshooting knowledge operators can question in plain language, on a tablet at the line, in English or Spanish.
- Faster onboarding and changeovers
- Institutional knowledge captured before it retires
ML demand forecasting and inventory optimization
Statistical and machine-learning forecasts by SKU and location, with safety-stock optimization that feeds MRP parameters directly.
- Better forecast accuracy and fill rate
- Less cash tied up in the wrong inventory
Digital twin and simulation
Discrete-event models of lines, plants, and networks to test layouts, capacity, crewing, and footprint scenarios before committing capex.
- De-risks expansions and consolidations
- Finds the real bottleneck before you buy equipment
Unified namespace and IT/OT convergence
One data architecture from PLC and MES to ERP and Power BI. Industrial DataOps so dashboards and models run on the same trusted numbers.
- Kills duplicate reports and reconciliation
- Foundation every other use case depends on
Agentic planning workflows
AI agents that draft the production schedule, flag supply risks, and prepare the S&OP pack, with planners approving instead of assembling.
- Planner time shifts from spreadsheets to decisions
- Faster response to demand and supply changes
Energy, sustainability, and OT security
Energy monitoring by line and shift, waste and carbon reporting from the same data model, and OT cybersecurity basics so connecting the plant doesn't expose it.
- Lower utility cost per case
- Audit-ready sustainability reporting
Where does your plant stand?
Twenty-five questions, five per discipline. Score yourself honestly and get a benchmark, a discipline-by-discipline analysis, and a 180-day roadmap. About eight minutes.
Maturity
Score by discipline
Strengths to build on
Biggest gaps
Discipline-by-discipline analysis
Recommended 180-day roadmap
Results stay in your browser and are not stored or sent anywhere.
Deep experience where margins are made on the floor.
Fifteen-plus years across food and beverage and discrete manufacturing, from high-speed bottling lines to one-off custom fabrication.
Food and beverage
High-volume, perishable, and regulated. Yield, changeovers, cold chain, and food safety decide the margin.
- Bottling
- Bakery
- Dairy
- Prepared foods
- Grocery
- Frozen
- Fresh
- DSD networks
Discrete and fabrication
Takt-paced assembly, fabrication, and finishing where labor balance, kitting, and quality at the source drive throughput.
- Assembly lines
- Sub-assembly
- Kitting
- Fabrication
- Welding
- Powder coating
- Upholstery
- Spare parts
Medical devices
Custom fabrication of medical wheelchairs: engineered-to-order work with traceability requirements and unforgiving lead-time expectations.
- Custom fabrication
- Engineer-to-order
- Traceability
- Service parts
Private equity and ownership transitions
Multiple plant turnarounds that carried private equity owners to exit, and due diligence support on the buy side. For sponsors and management teams, that means operational diligence, 100-day plans, and value-creation programs written by someone who has had to deliver them.
Multi-plant networks
Leading industrial engineering and CI across a network of more than a dozen plants, and P&Ls up to $2B with teams of 2,000+, means we know how to make improvement spread across a network instead of staying in the plant where it started.
Ambition in action
Problems we've taken on and what changed. Client names withheld.
Plant turnaround ahead of a private equity exit
A prepared foods plant was missing budget, losing customers on service, and dragging down the valuation of the business.
- Rebuilt daily management around OEE, yield, and labor per case
- Cut changeover time and scrap on the highest-volume lines first
- Reset crewing and shift patterns to the real demand curve
- Restored PM compliance to stop the reactive maintenance spiral
Sourcing reset across raw materials
A multi-plant manufacturer bought the same ingredients and packaging on different terms at every site, with contracts rolling over on autopilot.
- Consolidated spend by category and re-bid every major contract
- Standardized specs so suppliers could quote real volume
- Built a procurement dashboard for price, volume, and expiry
Supply and demand model and MRP reset
Planners overrode MRP by hand because the parameters were wrong: stale lead times, arbitrary safety stocks, lot sizes nobody had reviewed in years.
- Built a full supply and demand model from forecast to schedule
- Reset MRP parameters item by item, tied to real lead times and variability
- Launched cycle counting with root cause on every variance
- Stood up a monthly S&OP cadence with one set of numbers
ERP implementation without a stoppage
A growing manufacturer had outgrown its legacy system. Previous attempts stalled on dirty master data and a plant that couldn't afford downtime.
- Led selection, design, and go-live as the operations owner
- Cleaned item, BOM, and routing data before configuration
- Phased cutover by plant and function with parallel runs
- Built Power BI reporting on the new data model from day one
WMS rollout and live plant dashboards
A cold and dry storage operation ran on spreadsheets and radio calls. Leadership found out about problems a day late.
- Implemented a WMS across cold and dry storage without stopping shipments
- Built dashboards for throughput, inventory accuracy, and labor
- Moved the daily review to a live screen on the floor
Hub and freight strategy for a DSD network
A fresh and frozen DSD business had grown by acquisition into overlapping routes, half-empty trucks, and freight contracts nobody had renegotiated.
- Modeled the network and defined a hub strategy that shortened routes
- Re-bid freight and secured multi-year carrier contracts
- Redesigned frozen and fresh route structures separately
CI system across a multi-plant network
More than a dozen plants, each with its own way of measuring efficiency. Good ideas in one plant never reached the others.
- Set common labor standards and an IE playbook for the whole network
- Built one CI system with shared metrics and plant scorecards
- Trained plant-level CI leads and ran cross-plant kaizen events
Takt-based assembly for custom medical wheelchairs
A custom fabrication shop built every unit as a one-off. Welding, powder coat, upholstery, and assembly each waited on the last, and lead times drifted for weeks.
- Separated standard sub-assemblies from true custom work and moved them to kitted, takt-paced cells
- Balanced welding, powder coating, upholstery, and final assembly to one takt time
- Set up a spare parts flow so service orders stopped pulling from production
Find the constraint. Prove it. Fix it.
Every line has one bottleneck at a time. We measure each station against its true rate, model what happens when the constraint moves, and put the fix on a CAD layout your maintenance and capital teams can act on. Below is a working model built from a real bakery line.
Interactive line constraint model
Each station is drawn at its measured rate in units per minute. The slowest station sets the line. Tap a station for detail, then use the slider to test a fix and watch the constraint move downstream.
Capacity ladder (UPM)
Sheeter and end cutter follow, capped at 180 UPM
OEM tuning to 175, upgrade path toward 220
Belt speed and dwell study; tier addition above 220
The model uses the same logic we use on site: measured rate at each station, the constraint sets net output, and every upstream fix moves the constraint to the next-slowest station. Capital is only justified once the model shows the next constraint can absorb it.
Plant and line layouts you can build from
We draw the current state and the proposed state to scale: equipment footprints, conveyor paths, oven and freezer zones, operator positions, and material flow. Deliverables are DWG/DXF and PDF, ready for your maintenance team, contractors, and capital requests.
Current-state layout
Measured footprints, conveyor paths, and operator positions with material flow and travel distance called out.
Proposed-state layout
The fix drawn to scale: relocated equipment, new transfers, buffer sizing, and the capital list that goes with it.
Capacity model
Station-by-station rate model tied to the layout, so every capital request shows the throughput it buys.
Two weeks on the floor, one scorecard.
We score the plant on ten operating fundamentals, backed by time studies, rate checks, and a loss tree in dollars. The scorecard tells you where the money is; the loss tree tells you how much.
What we measure on site
- Rate checks at every station vs. nameplate and standard
- Time studies and crewing vs. engineered labor standards
- Downtime coding audit: are reasons real and actionable?
- Changeover observation and SMED opportunity
- Yield and giveaway sampling at fill, slice, or weigh points
- Material flow, travel distance, and WIP between stations
- Safety walk against OSHA fundamentals and near-miss data
- Tier meeting and visual management review
From reactive to predictive, in the right order.
We assess the maintenance system, not just the machines: criticality, PM content and compliance, planning and scheduling, spares, CMMS data quality, and skills. Then we sequence the path to condition-based and predictive maintenance on the assets that deserve it.
Asset criticality and health (sample)
| Asset | Criticality | PM compliance | MTBF (hrs) | MTTR (hrs) | Reactive % | Strategy |
|---|---|---|---|---|---|---|
| Makeup infeed gearbox | A | 71% | 210 | 4.2 | 58% | Rebuild + vibration |
| Case packer | A | 88% | 340 | 1.8 | 34% | PM redesign |
| Oven zone drives | A | 96% | 1,900 | 3.1 | 12% | Condition-based |
| Spiral freezer | A | 93% | 1,400 | 6.5 | 18% | Predictive (motor current) |
| Depositor | B | 90% | 620 | 1.1 | 22% | Operator care |
| Metal detectors | B | 100% | 4,800 | 0.5 | 4% | Run to schedule |
System review
Criticality ranking, PM task audit, planner/scheduler ratio, backlog health, spares coverage, and CMMS data quality.
PM and planning reset
Rewrite PMs on A assets, set weekly scheduling discipline, and get compliance above 90% with reactive work under 30%.
Condition-based and predictive
Vibration, thermal, and motor-current monitoring on the assets that dominate the downtime Pareto, feeding the dashboard.
Want this run on your line?
A line assessment takes one to two weeks on site and ends with the constraint model, layout, scorecard, and a ranked plan.
Points of view from the plant floor.
Short, practical, and written by someone who has had to make it work on a Monday morning.
Why most manufacturing AI pilots die in month four
The model was fine. The downtime codes were not.
Read
The pattern repeats. A vendor demo looks great, a pilot launches on one line, and four months later the plant manager quietly stops opening the dashboard. Nine times out of ten the model didn't fail; the inputs did. Downtime was coded "other" 40% of the time. The BOM had three versions. The PLC tag for the filler was mislabeled since 2019.
Before spending on AI, spend a month on reason codes, master data, and one clean data path from the line to a dashboard. Then run the model on a problem that already sits at the top of your Pareto. If the AI can't beat a good supervisor with a clean spreadsheet, it isn't ready, and neither is the plant.
Your MRP isn't broken, its parameters are
Why planners override the system, and how to make them stop.
Read
When planners override MRP every morning, leadership blames the software. Look at the parameters instead. Lead times were entered at go-live and never updated. Safety stocks are round numbers someone guessed. Lot sizes match a truck that no longer exists.
Reset them item by item, tied to measured lead-time variability and real service targets. Pair that with a cycle counting program so on-hand quantities can be trusted. Within a quarter the overrides stop, expedites drop, and the same system everyone hated starts running the plant.
The one number every plant should manage
OEE is a diagnostic, not a target.
Read
OEE is useful because it decomposes losses into availability, performance, and quality. It is a poor target because it can be gamed by running easy SKUs and hides the money. The number that should run the plant is the one that ties directly to the P&L: cost per case, cases per labor hour, or yield, depending on where the margin leaks.
Pick one. Put it on a screen at every tier meeting. Make one person accountable for it each week. Use OEE to explain why it moved, not as the goal itself.
What a 100-day plan should actually contain
Fewer initiatives, more measured losses.
Read
Most 100-day plans list twenty initiatives and deliver two. The ones that work start with a plant walk and a loss tree: where, in dollars, is the plant leaking margin today? Downtime, scrap, overtime, freight, inventory. Rank it. Take the top three.
Each gets an owner, a weekly number, and a dashboard by day 30. Everything else waits. Sponsors get a credible bridge to the value-creation case, and the management team gets early wins that fund the rest.
You can't predict a failure on a machine nobody maintains
Sequence PM discipline before predictive maintenance.
Read
Predictive maintenance is the most-cited AI use case in manufacturing and the most often skipped past. Sensors on a filler with 60% PM compliance will confirm what the mechanics already know: it's going to break. The sensor didn't add value; the missing PMs did the damage.
Get PM compliance above 90% and reactive work below 30% first. Then instrument the two or three assets that still dominate the downtime Pareto. That's where condition monitoring earns its keep.
The freight contract you haven't re-bid is costing you a plant
Distribution cost hides in plain sight.
Read
Manufacturers obsess over labor and materials and let freight ride on contracts signed years ago by someone who has since left. In DSD and cold chain especially, the gap between a modeled network and the inherited one is often worth more than a year of plant-floor kaizen.
Model the network. Separate fresh from frozen. Re-bid lanes with real volume data. Then fix the hub strategy so trucks leave full. The savings usually show up in the first quarter.
Plant Floor Notes
One short note a month on what's working in real plants: constraints, planning, dashboards, and AI that pays back. No selling.
Brandon Stone
Founder and Principal, Stone Consulting
Brandon is a manufacturing executive with 15+ years across plant operations, supply chain, distribution, technology, and engineering. He has led P&Ls from $50M to $2B and teams from ten people to more than 2,000, most recently as a Senior Director of Manufacturing.
He built his career on plant turnarounds, including multiple that carried private equity owners to exit, and has supported sponsors on due diligence. He has led industrial engineering and continuous improvement across a multi-plant network, lifted OEE by 8 to 25 points on constraint lines, solved bottlenecks from dough to case packer, saved 10%+ on raw material sourcing and contracts, set distribution and freight strategy, and delivered ERP and WMS implementations. As a Power BI developer with over 10 years of experience, he builds the live dashboards that keep every engagement honest.
Stone Consulting takes its name from the family and its structure from the five disciplines every plant depends on: Supply chain, Technology, Operations, Networks, and Engineering. The mark is a split hexagon: raw, faceted stone on one side and the machined part it becomes on the other. Raw material in, engineered result out.
Experience
- Senior Director of ManufacturingMulti-plant food manufacturing; P&L leadership from $50M to $2B, teams from 10 to 2,000+
- Industrial engineering and continuous improvement leaderMulti-plant network: bottleneck analysis, labor standards, CI system, plant scorecards; OEE gains of 8–25 pts
- Procurement and supply chain leaderNine-figure raw material accounts, 10%+ savings, supply and demand modeling, MRP strategy, cycle counting
- Distribution and logistics leaderFreight strategy, carrier contracts, hub strategy, cold and dry storage, WMS implementation
- Systems and analyticsERP implementations, WMS, Power BI developer with 10+ years of live dashboards
- Plant turnaround and private equityMultiple exits; buy-side due diligence and 100-day plans
Sector experience
Food and beverage: bottling, bakery, dairy, grocery, prepared foods, frozen and fresh DSD networks. Discrete: takt-based assembly, sub-assembly, kitting, fabrication, welding, powder coating, upholstery, spare parts, and custom medical wheelchair fabrication.
Based in
Cincinnati, Ohio. On site anywhere in North America.
Let's talk about your plant.
Book a 30-minute call, send a message, or submit an RFP. Brandon replies within one business day.
- EmailBrandon@stoneconsulting.org
- Phone509-539-9452
- LinkedInlinkedin.com/in/brandonstonemba
- Based inCincinnati, Ohio. On site anywhere in North America.
Common questions
Short answers to what most plant leaders and sponsors ask first.
How fast can you be on site?+
Usually within two weeks for a diagnostic. Brandon is based in Cincinnati and travels anywhere in North America.
Do you implement, or just advise?+
Implement. Every engagement has a number attached and Brandon stays on the floor until it moves. The deliverables are models, layouts, dashboards, and trained people, not a deck.
How are fees structured?+
Fixed scope and fixed fee for diagnostics and projects. For performance sprints with a measurable target, part of the fee is tied to the result.
What size of manufacturer do you work with?+
Single plants from about $30M in revenue up to multi-plant networks, plus private equity sponsors evaluating or owning them.
Which systems do you work in?+
Major ERPs, WMS and MES platforms, PLC and line data, and Power BI. Vendor-neutral: we have no reseller relationships.