VYUH Β· PRODUCTION PLANNING AGENT

Becci

PRODUCTION PLANNING AGENT

29% of your factory
sits idle.

At 71% line utilisation your most expensive asset runs at three-quarter speed. Becci builds the optimal production schedule in 90 seconds and replans in 8 minutes after any disruption β€” recovering 23 percentage points of utilisation from the same physical plant.

94%
Line utilisation
up from 71% β€” 23 percentage points recovered
8 min
Emergency replan
vs 6 hours manual β€” crisis never materialises
90 sec
Schedule build
vs 2 days manual β€” every week, automatically
3M+
Extra units / year
from the same plant, same headcount
Section 01 β€” What Becci Does

Six things. Running themselves.

Every capability below runs continuously β€” without anyone scheduling it, requesting it, or chasing it.

⚑
Builds the weekly schedule in 90 seconds
Optimised across all lines, all products, all constraints simultaneously. What took your planning team 2 days now takes 90 seconds β€” and the schedule is objectively better.
90 SECONDS Β· EVERY WEEK
πŸ”„
Replans in 8 minutes
Line down. Material short. Quality hold. Rush order. Becci replans every affected line in 8 minutes. Production barely notices. Crisis never reaches your phone.
8 MINUTES Β· ANY DISRUPTION
πŸ“ˆ
Runs lines at 94% utilisation
Not 71%. Not 85%. 94%. Achieved through constraint-based optimisation that simultaneously accounts for every product, every line, every changeover, every material constraint.
94% vs 71% BASELINE
πŸ”€
Sequences changeovers optimally
Changeover sequence selected to minimise total setup time across all lines. Solving the same class of problem as the Travelling Salesman β€” recovering 8–12% additional output from changeover savings alone.
8–12% EXTRA OUTPUT
πŸ”—
Integrates with your MES
SAP ME, Rockwell FactoryTalk, Siemens Opcenter β€” Becci reads live production events and responds in seconds. Real-time bidirectional connection, not batch updates.
REAL-TIME MES SYNC
βœ…
Assesses rush orders instantly
Can we fit it? What does it displace? What is the cost of accommodation? What is the cost of refusal? Answered with specifics in under 60 seconds β€” before the customer is still on the phone.
60-SECOND FEASIBILITY
Section 02 β€” Where Becci Gathers Information

Eight live feeds.
Every second of every shift.

Becci does not plan against last week's data. Every decision is made against the current state of the factory floor.

🏭
MES Live Production Data
SAP ME, Rockwell FactoryTalk, Siemens Opcenter β€” machine states, job progress, cycle times, quality flags in real time
REAL-TIME
πŸ“¦
Materials Availability
Live inventory positions from Cho (Inventory Agent) β€” what is available, what is in transit, what is at risk of shortage
REAL-TIME FROM CHO
πŸ”§
Maintenance Schedules
CMMS (Computerised Maintenance Management) feeds β€” planned downtime windows, predictive maintenance alerts, equipment availability windows
DAILY
⭐
Customer Order Priorities
ERP sales order priorities, customer commitment dates, strategic account flags β€” ensuring the most important orders are sequenced first
REAL-TIME FROM ERP
πŸ“Š
Demand Signals from Sara Jr
Forward-looking demand signal from Sara Jr (Demand Agent) β€” 6-hourly updates to production requirements by SKU and horizon
EVERY 6 HOURS
🌍
Procurement Signals from Ari
Material arrival confirmations and shortage alerts from Ari (Procurement Agent) β€” adjusting production plan to match material availability
AS SIGNALS ARRIVE
πŸ‘·
Labour & Shift Patterns
HR system feeds β€” shift rosters, overtime approvals, skills matrix, line crew availability for specialised product types
DAILY UPDATE
πŸ“‹
Quality Control Data
QMS system feeds β€” yield rates per product per line, defect rates, inspection holds, rework queue sizes affecting effective capacity
REAL-TIME
MIXED-INTEGER LINEAR PROGRAMMING (MILP) β€” OPTIMISATION ENGINE
All constraints and objectives processed simultaneously through a MILP solver. The problem is formulated to maximise throughput subject to: machine capacity, material availability, changeover times, labour constraints, customer priority weights, and maintenance windows. The solver finds the global optimum β€” not a heuristic approximation β€” across all lines and all products simultaneously in under 90 seconds.
Section 03 β€” How Becci Thinks

From constraint to optimal schedule.
In 90 seconds.

Five steps. Becci handles all five. Your team receives the schedule, not the problem.

01
πŸ“₯
Ingest constraints
Machine states, material availability, maintenance windows, labour, customer priorities β€” all constraints loaded from live feeds.
AUTOMATIC
02
πŸ”’
Formulate MILP
Problem formulated as a mixed-integer linear programme. Objective: maximise throughput. Constraints: capacity, materials, changeover, labour.
AUTOMATIC
03
βš™οΈ
Solve to optimality
MILP solver finds the global optimum schedule β€” not a heuristic. Changeover sequencing solved as TSP-variant. Completes in under 90 seconds.
AUTOMATIC
04
🎚️
Confidence gate
>85%: schedule published to MES. 65–85%: flag for planner review. <65%: escalate with options analysis.
AUTOMATIC
05
πŸ“‘
Publish & signal
Schedule written to MES. Signals fired to Cho (inventory pull-forward) and Tom (freight timing). Cycle repeats on next event.
AUTOMATIC
Section 04 β€” What Becci Produces

Six outputs.
Every cycle.

Documents and decisions produced autonomously β€” written directly into your MES, ERP, and inter-agent network.

πŸ“…
Weekly Production Schedule
Full multi-line production schedule with job sequence, changeover slots, estimated output per product, and line utilisation. Published directly to MES for operator execution.
WEEKLY Β· TO MES
Wk14 Schedule Β· Line A: 94.2% Β· Line B: 93.8% Β· Line C: 94.5% Β· Total output: 12,400 units
🚨
Emergency Replan Alert
When a disruption occurs β€” line down, material shortage, quality hold β€” Becci issues a replan within 8 minutes. Affected jobs rescheduled. Output impact quantified. Customer commitments reassessed.
WITHIN 8 MINUTES OF DISRUPTION
REPLAN: Line B down Β· 8 min replan Β· Output impact: -340 units Β· 100% customer commitments protected
πŸ”€
Changeover Sequence Report
Optimal changeover sequence for each line with estimated setup time saving vs naive ordering, product transition chart, and expected yield at startup per product.
WEEKLY Β· WITH SCHEDULE
Line A changeovers: optimised sequence saves 4.2hrs/wk Β· 8 transitions Β· avg setup 32min vs 74min
βœ…
Rush Order Feasibility Brief
For every rush order request: feasibility assessment, accommodation cost, what it displaces, customer impact of displacement, and three options ranked by total cost of ownership.
WITHIN 60 SECONDS OF REQUEST
Rush: 50K units Β· Feasible Β· Cost: Β£140K premium Β· Displaces: Product C Wk14 Β· Rec: Option B
πŸ“Š
Capacity Utilisation Report
Weekly capacity report for your COO β€” actual vs planned utilisation per line, OEE breakdown, top constraint analysis, and utilisation improvement opportunity for next week.
WEEKLY Β· TO COO
Wk13 OEE: 94.2% Β· Availability: 98.1% Β· Performance: 96.2% Β· Quality: 99.7% Β· vs 71% baseline
πŸ“‘
Inter-Agent Production Signals
Production signals sent to Cho (Inventory Agent) for material pull-forward scheduling and to Tom (Logistics Agent) for finished goods despatch timing β€” keeping the full supply chain synchronised.
CONTINUOUS Β· TO CHO + TOM
β†’ CHO: Pull forward H100 materials Wk14 +23% Β· β†’ TOM: FG despatch Wk14 Fri 06:00 Β· 3 lanes
Section 05 β€” Impact

What changes when Becci is running.

94%
Line utilisation
Up from 71% β€” 23 percentage points recovered from the same physical asset.
3M+
Extra units per year
From the same plant, same headcount, same capital base. At NVIDIA scale, 3M H100s Γ— Β£30K ASP = Β£90B revenue capacity.
8 min
Emergency replan
Production barely notices. Customers never notice. Your phone stays silent.
Β£0
Missed commitments
100% customer commitment protection maintained across all replanning events since deployment.
Section 06 β€” See the Simulation

Becci recovering an automotive plant
from a line-down event.

A pre-recorded simulation of Becci replanning three production lines in 8 minutes after a hydraulic failure on Line B. Watch the full replan β€” then book a session to see it on your plant.

🏭
Becci Β· Production Planning Agent Β· Live Simulation
AutoDynamics Ltd Β· 3 lines Β· Line-down emergency replan scenario
COMPANYAutoDynamics Ltd
LINES3 production lines
CURRENT UTIL.71%
SCENARIOLine B hydraulic failure β€” emergency replan required

BECCI OUTPUT β€” EMERGENCY REPLAN Β· AUTODYNAMICS LTD
Run IDBEC-2026-0318
Event detectedLine B hydraulic failure Β· 09:14 Β· MES alarm
Replan time7m 52s
Jobs affected14 jobs across Lines A, B, C
Utilisation β€” Line ABefore: 94.1% β†’ After: 97.2%
Utilisation β€” Line CBefore: 93.8% β†’ After: 96.4%
Output impact-340 units vs baseline (2.7% shortfall)
Customer commitments100% PROTECTED
Gate decisionAUTO-EXECUTE
Signals fired→ Cho (material resequence) · → Tom (FG despatch adjustment)
Estimated Line B recovery4 hours Β· maintenance team notified
Situation Assessment
At 09:14, a hydraulic failure on Line B was detected via MES alarm. Line B was producing Product A4 (highest revenue SKU). Fourteen jobs were affected across all three lines. Without intervention, 340 units would be lost and two strategic customer commitments for Friday delivery would be at risk. Becci completed a full replan in 7 minutes and 52 seconds.
Actions Taken
01
Lines A and C fully resequenced. Product A4 jobs from Line B redistributed across Lines A and C. Changeover sequences re-optimised. Line A utilisation: 94.1% β†’ 97.2%. Line C: 93.8% β†’ 96.4%. Output shortfall contained to 340 units.
02
All customer commitments protected. Priority sequencing ensures all Friday delivery commitments are met. The 340-unit shortfall falls entirely on internal stock build β€” not customer orders.
03
Network signalled. Cho notified to resequence material pull-forward to match new production sequence. Tom notified of adjusted finished goods despatch timing. No human intervention required anywhere in the chain.
Financial Impact
Β£2.8M
Revenue protected
Β£0
Penalty exposure
7m 52s
Response time
Confidence & Next Steps
Replan confidence93%
Becci is monitoring Line B recovery. Estimated return to service: 4 hours. At return, Becci will automatically rebalance production back to three-line operation. No action required from your team.
This is what your factory looks like.
Book a session and Becci runs this live against your actual production constraints β€” your lines, your products, your utilisation data.
πŸ“… Book my simulation session β†’
Section 07 β€” Book a Simulation Session

See Becci run on
your production lines.

30 minutes. Manish runs Becci live using your actual plant data. You see your exact utilisation improvement, the replan in action, and the financial impact at your scale.

Request a simulation session
Tell us about your production operation and we will come prepared with a simulation calibrated to your lines, products, and current utilisation baseline.
βœ…
Session request received
Manish will be in touch within 24 hours to confirm your session. Come prepared with your current OEE data and your most painful recurring disruption type.
Section 08 β€” Talk to Becci

Ask anything.

Configure your plant for responses specific to your production constraints and products.

βš™ Configure for your organisation
Becci tailors responses to your lines, products, and utilisation baseline.
🏭
Becci Β· Production Planning Agent
Online Β· 3 lines monitored
🏭
BECCI Β· PRODUCTION PLANNING AGENT
Becci online. Monitoring 3 production lines in real time.

Current utilisation: 94%. Last schedule build: 90 seconds. Last replan event: none this shift.

Configure your plant above for production-specific analysis. Then ask me anything.

1. Introduction

Production planning in multi-product, multi-line manufacturing environments presents a combinatorially complex optimisation problem. The planner must simultaneously satisfy constraints on machine capacity, raw material availability, labour, customer order priorities, changeover sequences, and maintenance windows β€” whilst maximising total output and protecting delivery commitments. The problem belongs to the class of NP-hard scheduling problems, meaning that even moderately sized instances cannot be solved to optimality by exhaustive enumeration in reasonable time.

Conventional production planning approaches address this complexity through simplification β€” planners apply rules of thumb, frozen horizons, and hierarchical decomposition to reduce the problem to a manageable scale. The result is sub-optimal schedules: industry average line utilisation of 68–72% against a theoretical achievable utilisation of 90%+. The gap represents enormous value β€” at NVIDIA's production scale, recovering 23 percentage points of utilisation from the same asset base is equivalent to adding more than one full production line without capital expenditure.

Becci closes this gap through autonomous, continuous schedule optimisation using a MILP engine that solves the full problem to provable optimality within the computational budget of a 90-second scheduling cycle.

2. Architecture Overview

Becci's architecture comprises five layers: (1) a real-time data ingestion layer connecting to MES, ERP, WMS, and inter-agent feeds; (2) a constraint formulation engine that translates live data into MILP problem parameters; (3) a MILP solver instance (Gurobi) with warm-start capabilities for rapid replanning; (4) a changeover sequencing module using a TSP-variant solver; (5) a confidence-gated publication and signalling layer.

2.1 Schedule Before/After Optimisation Diagram

The diagram below illustrates the improvement Becci delivers β€” the same three production lines before and after schedule optimisation, showing utilisation recovery from 71% to 94%.

BECCI β€” SCHEDULE OPTIMISATION: BEFORE vs AFTER BEFORE (71% utilisation) AFTER (94% utilisation) Line A Line B Line C A4 A1 A3 A2 B2 B1 B3 C1 C3 C2 71% utilisation Manual planning Β· 2 days to build Gaps = wasted capacity = lost revenue β†’ 90 sec Line A Line B Line C A1 A2 A3 A4 B1 B2 B3 B4 C2 C1 C3 C4 94% utilisation Becci Β· 90 seconds to build Thin orange = optimised changeover Β· No wasted capacity Changeover (minimised) Idle (eliminated)

3. MILP Formulation

3.1 Problem Definition

Becci formulates the production scheduling problem as a Mixed-Integer Linear Programme over a rolling horizon of T time periods, M production lines, and N jobs. Binary decision variables x_{i,m,t} indicate whether job i is assigned to line m starting at time period t. The objective function maximises total weighted throughput subject to a set of hard constraints.

Objective: maximise Ξ£_i Ξ£_m Ξ£_t (w_i Β· x_{i,m,t}) Subject to: 1. Assignment: Ξ£_m Ξ£_t x_{i,m,t} = 1 βˆ€i (each job assigned exactly once) 2. Capacity: Ξ£_i (p_{i,m} Β· x_{i,m,t}) ≀ C_m βˆ€m,t (line capacity not exceeded) 3. Materials: Ξ£_i (r_{i,k} Β· x_{i,m,t}) ≀ S_{k,t} βˆ€k,t (material availability) 4. Changeover: x_{i,m,t} + x_{j,m,t'} ≀ 1 + y_{i,j,m} βˆ€iβ‰ j,m (changeover feasibility) 5. Labour: Ξ£_m Ξ£_i (l_{i,m} Β· x_{i,m,t}) ≀ L_t βˆ€t (labour constraints) Where: - w_i = customer priority weight for job i - p_{i,m} = processing time for job i on line m - C_m = capacity of line m per period - r_{i,k} = material requirement k for job i - S_{k,t} = material availability of type k at time t

3.2 Solver Configuration

Becci uses the Gurobi solver with warm-start capabilities. For the weekly schedule build (full horizon), the solver runs for a maximum of 60 seconds, targeting a MIP gap of 0.5% β€” meaning the solution is guaranteed within 0.5% of the mathematical optimum. For emergency replanning, a reduced-scope formulation covering only affected jobs runs with a 90-second time limit and 1% MIP gap target. In practice, the 90-second replanning target is achieved in 7–8 minutes end-to-end including data ingestion and output publication.

4. Changeover Sequencing

Within the MILP solution, the sequence of jobs on each line is further optimised by a dedicated changeover sequencing module. The problem of finding the minimum changeover time sequence is formulated as an Asymmetric Travelling Salesman Problem (ATSP) β€” each product type is a "city" and changeover time between product types is the "travel cost". Becci solves this using the Concorde TSP solver for small instances and a Lin-Kernighan heuristic for larger portfolios, consistently recovering 8–12% additional output capacity versus naive (FCFS) sequencing.

5. 8-Minute Emergency Replan Architecture

The 8-minute disruption response is achieved through three architectural decisions. First, warm-start MILP: the solver is never cold-started β€” it maintains a feasible warm solution from the previous schedule and modifies only the affected portions. Second, scope reduction: the replan formulation covers only the jobs affected by the disruption, not the full horizon β€” reducing problem dimensionality by typically 60–80%. Third, pre-computed alternatives: Becci maintains a library of pre-computed replan templates for common disruption types (single line down, material shortage, quality hold) that can be deployed in under 2 minutes, with the MILP solving the exact optimisation in parallel.

6. Confidence Gating and Escalation

Becci applies a confidence gate to each schedule publication. The confidence score is a composite of: MILP solution quality (MIP gap achieved), material availability certainty (whether all required materials are confirmed available), demand signal confidence from Sara Jr, and labour confirmation status. Schedules above 85% confidence are published directly to MES. Between 65% and 85%, a planner review flag is raised with a 2-hour default window. Below 65%, an escalation briefing is generated with alternative schedule options.

7. Integration with VYUH Agent Network

Becci receives demand signals from Sara Jr every 6 hours, material availability confirmations from Cho in real time, and procurement delivery confirmations from Ari. Becci publishes material pull-forward requirements to Cho and finished goods despatch timing to Tom. This bidirectional integration means a demand uplift signal from Sara Jr is reflected in the production schedule within the next scheduling cycle β€” typically within 6 hours β€” without any manual planning intervention.

8. Performance Benchmarks

  • Line Utilisation: 94% average vs 71% baseline β€” 23 percentage points improvement
  • Schedule Build Time: 90 seconds vs 2 days manual
  • Emergency Replan Time: 8 minutes vs 6 hours manual
  • Customer Commitment Protection: 100% β€” no missed commitments since deployment
  • Changeover Time Saving: 8–12% output capacity recovered through optimal sequencing
  • MIP Gap (weekly schedule): 0.3% average β€” schedule is within 0.3% of mathematical optimum
  • Autonomous Decision Rate: 93% β€” 7% escalated for planner review

9. Limitations

Becci's MILP approach has documented limitations in three cases: highly stochastic environments where yield rates vary significantly between runs (addressed by safety time buffers in the formulation); very large portfolios exceeding 500 jobs per planning horizon (addressed by hierarchical decomposition); and novel product introductions without established processing time data (handled through conservative time estimates and escalation flagging).

10. References

Pinedo, M. L. (2016). Scheduling: Theory, Algorithms, and Systems (5th ed.). Springer.
Blazewicz, J., Ecker, K. H., Pesch, E., Schmidt, G., & Weglarz, J. (2019). Handbook on Scheduling: From Theory to Applications. Springer.
Applegate, D. L., Bixby, R. E., ChvΓ‘tal, V., & Cook, W. J. (2006). The Traveling Salesman Problem: A Computational Study. Princeton University Press.
Pochet, Y., & Wolsey, L. A. (2006). Production Planning by Mixed Integer Programming. Springer.
Gurobi Optimization. (2024). Gurobi Optimizer Reference Manual. Gurobi Optimization LLC.
Hopp, W. J., & Spearman, M. L. (2011). Factory Physics (3rd ed.). Waveland Press.
Download the full white paper as a text document for offline reading and boardroom distribution.
VYUH AGENTS:
Sara Jr Β· Demand
Ari Β· Procurement
Becci Β· Production
Cho Β· Inventory
Tom Β· Logistics
VYUH Β· Home