VYUH SUPPLY CHAIN NEURAL NETWORK · AGENT 04

CHO

INVENTORY MANAGEMENT AGENT
Your inventory knows exactly what it needs.
Before the shelves tell you.

Cho manages inventory across every warehouse, every SKU, every echelon — simultaneously. Multi-echelon optimisation calculates the precise safety stock needed at every node, frees trapped working capital without ever risking a stockout, and reorders autonomously with 99.97% accuracy.

99.97%
Inventory Accuracy
Across all warehouses, all SKUs, in real time
22 days
Optimised Safety Stock
Down from 90-day fat buffers industry average
£600M
Working Capital Freed
Released from excess inventory without stockout risk
2,400
SKUs Managed
Every reorder point recalculated daily
What Cho Does

Six ways Cho keeps your inventory exactly right

Cho eliminates both stockouts and excess. Not one or the other — both.

🏗️
Multi-echelon inventory optimisation
Cho models your entire inventory network — central DC, regional hubs, forward positions — and calculates the globally optimal stock allocation across every echelon simultaneously. Not each warehouse independently. All of them together.
MULTI-ECHELON · GLOBAL OPTIMUM
📊
Dynamic safety stock calculation
Traditional safety stock is set once in a spreadsheet and forgotten. Cho recalculates it every 24 hours using real demand variability, supplier lead time distributions, and service level targets. Your buffer is always exactly right — not a 1998 heuristic.
RECALCULATED DAILY · DEMAND-DRIVEN
Autonomous reorder triggering
When inventory crosses the reorder point, Cho raises a reorder signal to Ari — confirmed quantity, timing, supplier preference, and urgency level. No spreadsheet. No planner approval required for routine replenishment. 91% of reorders execute without human touch.
91% AUTONOMOUS · SIGNALS TO ARI
🔴
Stockout risk forecasting
Cho monitors lead time risk from Ari, demand surge signals from Sara Jr, and current stock levels — calculating a 14-day stockout probability for every critical SKU. High-probability items trigger alerts and pre-emptive reorders before the shelf empties.
14-DAY HORIZON · PROBABILISTIC
♻️
Obsolescence and expiry management
Cho tracks batch tracking, expiry dates, and demand trajectories together. Items approaching obsolescence are flagged early — with markdown recommendations, redistribution options across warehouses, and automatic reduction in future reorder quantities.
BATCH-LEVEL · EXPIRY TRACKING
🔄
Warehouse rebalancing
When Cho detects imbalance across locations — excess in one DC, shortage risk in another — it generates inter-warehouse transfer instructions automatically. Stock moves to where it is needed before it becomes a problem. Tom coordinates the freight.
INTER-WAREHOUSE · COORDINATES WITH TOM
Where Cho Gathers Information

Eight data streams feeding into every decision

Cho never relies on a single number. It cross-validates every signal across eight independent streams.

🏭
WMS Real-Time Feeds
Live warehouse management system data — stock on hand, location, movements, receipts, and picks across every DC and site.
LIVE · EVERY TRANSACTION
📡
RFID & IoT Sensors
Physical inventory verification via RFID scanning and IoT shelf sensors. Cho reconciles WMS records against physical counts continuously.
PHYSICAL VERIFICATION · CONTINUOUS
📦
GS1 Barcode Scans
Every inbound and outbound movement recorded at barcode scan. Cho uses this to verify receipt accuracy and update inventory positions instantly.
INBOUND + OUTBOUND · EVERY SCAN
Expiry & Batch Tracking
Batch-level tracking of all perishable and date-sensitive SKUs. Expiry windows, FEFO rotation logic, and markdown triggers all managed by Cho.
BATCH-LEVEL · FEFO LOGIC
📈
Sara Jr Demand Forecasts
Cho receives the full 30/60/90/180-day demand signals from Sara Jr, using them to project forward inventory positions and calculate safety stock requirements.
AGENT SIGNAL · 4× DAILY
🛒
Ari Reorder Confirmations
When Ari confirms a purchase order, Cho updates the in-transit inventory position immediately. Lead time estimates from Ari feed directly into Cho's stockout risk model.
AGENT SIGNAL · ON EVERY PO
🗺️
Storage Capacity Data
Warehouse capacity constraints, hazardous material segregation rules, temperature zone availability, and velocity-based slotting data inform Cho's allocation decisions.
CAPACITY · SLOTTING · ZONES
💹
Carrying Cost & Capital Data
Category-level holding costs, capital cost of inventory, and working capital targets feed into Cho's optimisation objective — balancing service level against financial efficiency.
FINANCIAL · FROM FINANCE SYSTEMS
MULTI-ECHELON INVENTORY OPTIMISATION ENGINE
Cho formulates inventory allocation as a stochastic multi-echelon optimisation problem. The objective function simultaneously minimises total holding cost and stockout risk across all nodes in the network, subject to capacity constraints, service level targets per SKU category, and lead time distributions. The solver handles up to 2,400 SKUs across 12 warehouse nodes simultaneously — recalculating the globally optimal allocation every 24 hours and triggering reorders in real time.
How Cho Thinks

Five-step inventory intelligence cycle

01
📥
Ingest & reconcile
All WMS feeds, RFID data, and GS1 scans reconciled into a single verified inventory position per SKU per location. Discrepancies flagged immediately.
AUTOMATIC
02
🔢
Multi-echelon optimise
Stochastic multi-echelon model runs daily. Demand variability and lead time distributions from Sara Jr and Ari feed the objective function. Globally optimal safety stock computed for every node.
AUTOMATIC
03
📊
Reorder point comparison
Real-time inventory position vs dynamic reorder point. Continuous monitoring — when the gap closes, a reorder signal is queued. Urgency level calculated from stockout probability and supplier lead time.
AUTOMATIC
04
🚦
Confidence gate
Reorder signals above 85% confidence execute autonomously to Ari. Signals between 65–85% flag for planner review. Below 65%, an escalation briefing is generated with full context and options.
AUTOMATIC
05
📤
Publish & signal
Reorder signals sent to Ari. Stockout risk alerts published to procurement and planning teams. Inter-warehouse transfer instructions generated. Tom notified of any urgent freight requirements.
AUTOMATIC
What Cho Produces

Six outputs that land in your teams every day

Inventory Accuracy Report
Daily reconciliation of WMS records vs physical verification across all 12 warehouses. Discrepancy rate, root cause flags, and recommended cycle count schedule.
DAILY · TO WAREHOUSE OPERATIONS
Accuracy: 99.97% · 2,400 SKUs · 12 locations · 3 discrepancies flagged for investigation
🛒
Autonomous Reorder Signal
Structured reorder signal to Ari with SKU, quantity, preferred supplier, required delivery date, and urgency classification. 91% of routine replenishment signals execute without human review.
CONTINUOUS · TO ARI PROCUREMENT AGENT
REORDER: PCIe Gen5 Conn. · Qty: 4,800 · Supplier: preferred · RDD: 14 days · Urgency: routine
📊
Safety Stock Recommendation
Monthly report showing the full optimised safety stock levels for every SKU, with rationale (demand variability, lead time risk, service level target), and comparison to current holdings.
MONTHLY · TO SUPPLY CHAIN DIRECTOR
Category avg: 22 days (was 90) · £600M released · Service level maintained: 99.1%
🔴
Stockout Risk Alert
Real-time alert when Cho calculates a stockout probability above 15% within the 14-day horizon. Includes current stock, demand projection, lead time status, and pre-emptive reorder recommendation.
REAL-TIME · TO PROCUREMENT + PLANNING
RISK ALERT: RTX 4090 GPU · P(stockout 14d): 34% · Pre-emptive reorder queued · Awaiting confirmation
⚠️
Obsolescence Flag
Weekly report identifying SKUs with excess inventory relative to forecast consumption, approaching expiry, or demand trend decline. Includes markdown recommendation, redistribution options, and future order reduction.
WEEKLY · TO COMMERCIAL + OPS TEAMS
FLAG: DDR4 ECC 32GB · Excess: 12,000 units · 94-day cover vs 22-day target · Recommend: markdown 15%
🔄
Warehouse Rebalancing Instruction
When stock imbalance is detected across network nodes, Cho generates a rebalancing instruction — move quantities, origin DC, destination DC, timing. Tom coordinates the inter-site freight automatically.
AS REQUIRED · TO TOM LOGISTICS AGENT
REBALANCE: NVMe SSDs · Move: 3,200 units · From: Manchester DC · To: Frankfurt DC · ETA: 3 days
Impact

What Cho changes in the first 90 days

£600M
Working capital freed
Released from excess safety stock and obsolete inventory. Reinvested into growth, not buffer.
99.97%
Inventory accuracy
Up from an industry average of 95.3%. Every order, every shipment, every pick — correct.
75%
Stockout reduction
Pre-emptive reorders eliminate the emergency premium freight and lost sales from stock-outs.
91%
Autonomous replenishment
Nine in ten routine reorders execute without a planner. Your team handles exceptions, not routine.
See the Simulation

Watch Cho free £180M in 4 minutes

A pre-recorded simulation of Cho analysing 12 warehouses across 2,400 SKUs and identifying £180M of working capital locked in excess safety stock — then generating the phased release plan. This is what your inventory would look like.

📦
Cho · Inventory Optimisation Simulation
Safety stock reduction scenario · GlobalTech Distribution · 12 DCs · 2,400 SKUs
SCENARIO
Safety stock reduction · £180M working capital release
COMPANY
GlobalTech Distribution
SCOPE
12 DCs · 2,400 SKUs · £840M total inventory
DURATION
24-second animation · 4 minutes real time
▸ CHO OPTIMISATION OUTPUT — GLOBALTECH DISTRIBUTION
SKUs analysed2,400 across 12 DCs
Excess inventory identified847 SKUs over-stocked vs optimal
Current avg safety stock88 days (vs 22-day optimum)
Working capital locked£180M in excess buffer
Reorder signals queued23 to Ari (routine) · 3 escalated
Service level impact99.1% maintained
Phased release plan£45M/month over 4 months
Confidence score93.2% — autonomous execution
Situation Assessment
Cho completed a multi-echelon analysis of GlobalTech Distribution's entire inventory network — 2,400 SKUs across 12 distribution centres. The analysis found that 847 SKUs are held significantly above their mathematically optimal safety stock level. The average safety stock across the network is 88 days of cover, against an optimal level of 22 days calculated from real demand variability and supplier lead time distributions. The excess represents £180M of working capital locked in unnecessary buffer — with no service level benefit.
Actions Taken
01
Phased release plan generated. £180M released over 4 months in equal tranches of £45M — allowing operations to adjust without disruption. Release sequenced by SKU risk level.
02
23 reorder signals sent to Ari. Revised reorder points replace previous over-stocked levels. All routine signals execute autonomously. 3 high-value SKUs escalated for commercial review.
03
Obsolescence flags raised for 14 SKUs. Items with excess beyond the release horizon flagged for markdown or inter-warehouse redistribution. Tom has been briefed on 6 transfer movements.
Financial Impact
£180M
Working capital released
99.1%
Service level maintained
£12.6M
Carrying cost saving / year
Recommendation
Cho will implement the phased release plan over 4 months, monitoring service levels continuously. If any SKU shows stockout probability rising above 10% during the transition, the release for that SKU will pause automatically. No service level risk exists at current confidence levels (93.2%). Cho is monitoring 3 escalated SKUs pending your review in the dashboard.
"This is what your inventory network would look like. Book a session and we run it live on your data."
📅 Book my simulation session →
Book a Session

See Cho run on your inventory

Request a simulation session
Tell us about your inventory operation and we will come prepared with a simulation calibrated to your warehouse network, SKU base, and working capital profile.
Session requested
We will be in touch to confirm a time. Cho will be ready for your data.
Talk to Cho

Ask Cho about your inventory

Configure your operation below and every response becomes specific to your warehouse network, SKU base, and supply chain.

📦
CHO
Inventory Management Agent · Online
✓ Configured
📦
CHO
I manage inventory across your full warehouse network — every SKU, every location, every reorder point. I can tell you where your working capital is trapped, which SKUs are at stockout risk in the next 14 days, and what your optimal safety stock levels should be. Configure your operation above for specific analysis, or ask me anything now.
VYUH Technical White Paper — Cho
Multi-Echelon Inventory Optimisation at Scale: The Cho Autonomous Inventory Management Agent
VYUH Supply Chain Neural Network · Technical Research Series · 2026
Inventory Systems Architecture · Applied Operations Research
ABSTRACT

This paper presents Cho, the autonomous inventory management agent within the VYUH Supply Chain Neural Network. Cho employs a stochastic multi-echelon inventory optimisation model to calculate globally optimal safety stock levels and reorder points across complex distribution networks. In deployment across a 12-DC, 2,400-SKU electronics distribution network, Cho has reduced average safety stock from 88 days to 22 days, released £600M in working capital, maintained 99.97% inventory accuracy, and achieved 91% autonomous replenishment execution. We describe the full technical architecture, mathematical formulation, confidence gating mechanism, and inter-agent coordination framework.

SAFETY STOCK DISTRIBUTION — BEFORE VS AFTER CHO 0 22 60 90 120+ DAYS OF SAFETY STOCK 0 25% 50% 75% OPTIMAL 22 days WAS 88 days £180M working capital freed area between curves = trapped cash Before Cho After Cho

1. Introduction

Inventory management in complex distribution networks presents a fundamental tension between service level and working capital efficiency. Traditional approaches — whether spreadsheet-based reorder point models or first-generation ERP replenishment modules — address this tension through static safety stock calculations, periodic review cycles, and conservative buffer policies. The result is predictable: excess inventory accumulates to protect against uncertainty, working capital is trapped at scale, and stockouts still occur because the static models cannot respond to real-time demand and supply signals.

The typical electronics distribution company holds 60–90 days of safety stock across its network. Academically optimal levels — calculated from actual demand variability and supplier lead time distributions — are typically 18–25 days. The gap between these numbers represents a substantial fraction of a company's working capital, earning no return and degrading the balance sheet.

Cho closes this gap through autonomous, continuous multi-echelon optimisation — recalculating optimal inventory levels daily using live signals from across the VYUH agent network and executing replenishment decisions autonomously for all routine cases.

2. Architecture Overview

Cho's architecture comprises four integrated layers: (1) a real-time data ingestion layer connecting to WMS, RFID sensors, GS1 scan systems, and inter-agent feeds from Sara Jr and Ari; (2) a multi-echelon optimisation engine that formulates and solves the network-wide inventory allocation problem daily; (3) a real-time monitoring and triggering layer that continuously compares live inventory positions to dynamic reorder points and queues reorder signals; (4) a confidence-gated publication layer that routes autonomous signals to Ari and escalates exceptions for human review.

The system maintains a live digital twin of the entire inventory network — every SKU, every location, every in-transit quantity, every pending reorder. This twin is updated on every WMS transaction, every Ari PO confirmation, and every Sara Jr forecast cycle — ensuring the optimisation engine always operates on current data.

3. Multi-Echelon Optimisation Formulation

Cho formulates the inventory allocation problem as a stochastic multi-echelon optimisation. The network is modelled as a directed graph where nodes represent inventory holding locations (central DC, regional DCs, forward positions) and edges represent replenishment flows with associated lead time distributions.

For each SKU i at location j, Cho calculates the optimal base stock level S*ᵢⱼ that minimises the expected total cost:

Minimise Σᵢⱼ [hᵢⱼ · E[Iᵢⱼ] + bᵢⱼ · E[Bᵢⱼ]]

Subject to: service level constraints per SKU category, network flow conservation, capacity constraints per location, and budget constraints on total inventory investment. Here hᵢⱼ is the holding cost per unit-day, bᵢⱼ is the stockout penalty, E[Iᵢⱼ] is the expected on-hand inventory, and E[Bᵢⱼ] is the expected backorders.

Demand at each node is modelled as a compound Poisson process with parameters estimated from the most recent 24 months of historical data and updated with Sara Jr's real-time demand signals. Lead time distributions are maintained per supplier per SKU from Ari's confirmed delivery history, updated on every completed purchase order.

4. Dynamic Safety Stock Calculation

The safety stock at any node is derived from the base stock level minus the expected demand during lead time: SS = S* - μL, where μL is the expected demand during the lead time. The key innovation relative to traditional methods is that both S* and μL are recalculated daily — incorporating current demand variability from Sara Jr and current lead time distributions from Ari — rather than set once from historical averages.

This dynamic recalculation is the mechanism through which Cho reduces safety stock without increasing stockout risk. As Sara Jr improves forecast accuracy over time, the demand variability estimate decreases — allowing safety stock to shrink safely. As Ari negotiates more reliable supplier lead times, the lead time variance decreases — allowing further safety stock reduction. The 22-day figure is not a target; it is the current mathematical optimum given current forecast accuracy and current supplier reliability.

5. Confidence Gating and Autonomous Execution

Cho applies a composite confidence score to each reorder signal before execution. The confidence score integrates: forecast confidence from Sara Jr (how certain is the demand signal driving this reorder?), supplier lead time reliability from Ari (how consistent is this supplier's lead time?), inventory data freshness (how recently was this SKU's count physically verified?), and reorder quantity normalcy (is this quantity within the normal range for this SKU, or is it anomalously large?).

Signals above 85% confidence execute autonomously to Ari. This threshold is calibrated to deliver 91% autonomous execution — meaning the team handles approximately 9% of reorders that involve some uncertainty. 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 reorder quantities and context.

6. Inter-Agent Coordination

Cho operates as a tight node within the VYUH agent network. Sara Jr's demand forecasts arrive every 6 hours and update Cho's demand models — immediately propagating any forecast revision into the inventory position projections and reorder point calculations. Ari's purchase order confirmations update Cho's in-transit inventory position in real time, preventing duplicate reorders. Tom receives inter-warehouse transfer instructions from Cho for rebalancing movements, coordinating the freight logistics autonomously.

This inter-agent architecture eliminates the information latency that causes most inventory management failures. In traditional environments, a demand forecast change takes days to propagate into a revised reorder decision. In the VYUH network, it propagates in minutes — before the inventory position has time to deteriorate.

7. Results

In 90-day deployment at a 12-DC, 2,400-SKU electronics distribution network: safety stock reduced from 88 days average to 22 days (75% reduction). Working capital released: £600M. Inventory accuracy improved from 95.3% to 99.97%. Autonomous replenishment rate: 91%. Stockout frequency reduced by 75%. Carrying cost reduction: £12.6M per annum.

The working capital release follows a characteristic phased pattern: the first 30 days deliver the largest single tranche as the most obviously over-stocked SKUs are identified and release plans initiated. Months 2–4 deliver progressively refined optimisation as the demand models improve with additional data.

8. Limitations and Future Work

The current implementation has documented limitations in three areas: highly promotional environments where demand exhibits extreme short-term spikes that the Poisson model underestimates; networks with extremely long tail SKUs (99th percentile demand velocity very low) where sparse data limits the accuracy of demand variability estimates; and novel product introductions where there is insufficient historical data to calibrate lead time distributions. Each limitation is handled through conservative fallback parameters and escalation flagging.

Future development priorities include: integration of external demand sensing data (social media, competitor stockouts) into the demand model; dynamic service level targets that adjust by SKU profitability; and a multi-period optimisation formulation that jointly optimises ordering costs with holding costs across the full planning horizon.

VYUH Supply Chain Neural Network · Cho · Inventory Management Agent
Sara Jr Ari Becci Tom