Lighting factory scheduling and inventory forecasting: no more overstock or stockouts in peak seasons
Client Profile
Pain Points
Sharp seasonal swings
Orders flood in during peak season while lines idle in the off-season.
Inventory squeezed both ways
Raw materials either tie up capital or cause line stoppages.
Delivery promises are guesses
Sales promise deadlines by feel, leading to penalties.
Live Demo
How We Built It
Demand forecasting
Models over order history and market rhythm forecast 8-week demand by category.
Smart scheduling
Orders, capacity, molds and changeover time feed an optimal scheduling model.
Safety-stock advice
Dynamic reorder points per material based on demand volatility and lead time.
Not betting on one AI — combining the right few
For a real business scenario, we don't lock in a single technical path up front. We evaluate candidate paths on the same benchmark, then design a hybrid so each path does what it's best at.
General LLM API
Public-cloud models such as GPT-4o / Qwen / DeepSeek
- ✓Out-of-the-box, minimal integration cost, cheap to start
- ✕Limited generalization to vertical domains, unstable field-level accuracy, separate compliance design
Industry fine-tuned model
Domain fine-tuning of open-source models on industry data
- ✓Grasps industry semantics, much higher accuracy, private deployment
- ✕Needs labeled data plus ongoing training and upkeep
RAG + rule engine
Retrieval-augmented knowledge plus business rule checks
- ✓Highly targeted, explainable, best latency and cost
- ✕Requires continuous data-loop investment
Fusion, not selection
What we deliver is not a single model but a fused pipeline of primary path + semantic correction + graceful fallback: routine cases are handled by the industry model in seconds; low-confidence fields are re-understood by an LLM; on service anomalies the system degrades to a general API and flags manual review — achieving accuracy, cost, latency and controllability at once.
An explainable, degradable, evolvable AI pipeline
We treat AI engineering as a pipeline, not a black box: every step has clear responsibilities, inputs/outputs and fallbacks. The five-step pipeline plus four-layer architecture below is the technical skeleton.
Ingestion & access layer
Multi-channel access with unified auth, idempotency and masking
AI inference layer
Multi-path inference plus a confidence router switching by features
Business orchestration layer
Connecting results to real business flows with rule checks and closure
Data loop layer
Every online error flows back; models iterate monthly
Tech Stack
Delivery Process
Data audit
Organized two years of order and inventory data.
Modeling
Trained forecasting and scheduling models and validated accuracy.
Build
Built scheduling and inventory dashboards linked to ERP.
Pilot
One month of parallel run against the old method.
Switch
Full switch to AI scheduling with monthly calibration.
Before vs After
“Scheduling went from six hours a week to half an hour, and inventory turned almost a month faster — much less capital pressure.”
Where It Fits
Related Cases
AI Inspection & Smart Scheduling Assistant
Deployed an AI inspection and scheduling assistant for a small hardware factory, enabling smart upgrades at low cost.
DongguanAI Surface Defect Detection System
Deployed AI surface defect detection to replace manual inspection, sharply cutting missed defects and rework.
HuizhouPredictive Maintenance Platform
Built a predictive maintenance platform to shift from reactive repair to early warning, cutting downtime losses.