# Tensor Analytics

> GrepEye is the operational intelligence platform most ERPs don't have. ML demand forecasting, S&OP, supply netting, AI agents, and audit trails for manufacturing and supply chain practitioners.

## Site index

- [llms.txt](https://tensoranalytics.ai/llms.txt): Concise site summary for LLMs
- [sitemap.xml](https://tensoranalytics.ai/sitemap.xml): Machine-readable URL index
- [robots.txt](https://tensoranalytics.ai/robots.txt): Crawler policy

## Pages

- [Home](https://tensoranalytics.ai/): Tensor Analytics and GrepEye overview.
- [GrepEye Platform](https://tensoranalytics.ai/grepeye): Eight capability pillars on one operational data model.
- [Security & Trust Center](https://tensoranalytics.ai/security): RBAC, audit trails, encryption, compliance roadmap, and data handling.
- [Solutions](https://tensoranalytics.ai/solutions): Data infrastructure, ML forecasting, S&OP console, rules and AI agents.
- [Industries](https://tensoranalytics.ai/industries): Discrete manufacturing, pharma, food and beverage, consumer goods.
- [Work](https://tensoranalytics.ai/work): S&OP cycle reduction, safety stock optimization, tool consolidation.
- [prompt/ed](https://tensoranalytics.ai/prompt-ed): Knowledge base, cookbook, learning track, and community.
- [prompt/ed Cookbook](https://tensoranalytics.ai/prompt-ed/cookbook): Step-by-step planning recipes with worked examples.
- [prompt/ed Learning Track](https://tensoranalytics.ai/prompt-ed/learn): Structured course on planning foundations for manufacturing.
- [About](https://tensoranalytics.ai/about): Tensor Analytics mission, team, and principles.
- [Contact](https://tensoranalytics.ai/contact): Book a 45-minute GrepEye walkthrough.
- [Privacy Policy](https://tensoranalytics.ai/privacy): How Tensor Analytics handles personal data.
- [Terms of Service](https://tensoranalytics.ai/terms): Terms for using tensoranalytics.ai and GrepEye.
- [Cookie Policy](https://tensoranalytics.ai/cookies): Cookie and tracking technology policy.

## Content pillars

- [Demand forecasting](https://tensoranalytics.ai/prompt-ed/pillar/forecasting): Articles about demand forecasting.
- [S&OP & IBP](https://tensoranalytics.ai/prompt-ed/pillar/sop): Articles about s&op & ibp.
- [Inventory & supply](https://tensoranalytics.ai/prompt-ed/pillar/inventory): Articles about inventory & supply.
- [AI in operations](https://tensoranalytics.ai/prompt-ed/pillar/ai-ops): Articles about ai in operations.
- [Data & integration](https://tensoranalytics.ai/prompt-ed/pillar/data): Articles about data & integration.
- [The AI Hype Hangover](https://tensoranalytics.ai/prompt-ed/pillar/hype-hangover): Articles about the ai hype hangover.

## Articles

Each article is practitioner-focused content about demand planning, S&OP, inventory, AI in operations, and data infrastructure.

### Module 8: Audit, compliance, and the board pack
- URL: https://tensoranalytics.ai/prompt-ed/module-8-audit-compliance-board-pack
- Type: module | Pillar: AI in operations | Date: 2026-07-08
- Author: Kislay S | Read time: 13 min
- Tags: Learning track, Foundations, Audit trail, Compliance, Board pack
- Summary: The final module of the Planning Foundations course. How to close the planning cycle with a defensible record. Audit trails, regulatory compliance, and the board pack that communicates the plan to leadership.
### Module 7: AI agents and human-in-loop
- URL: https://tensoranalytics.ai/prompt-ed/module-7-ai-agents-human-in-loop
- Type: module | Pillar: AI in operations | Date: 2026-07-01
- Author: Kislay S | Read time: 12 min
- Tags: Learning track, Foundations, AI agents, Human-in-loop, Automation
- Summary: The seventh module of the Planning Foundations course. How AI agents assist with the planning cycle, what human-in-loop actually means, and why the agent proposes and the planner disposes.
### Module 6: Scenario planning and what-ifs
- URL: https://tensoranalytics.ai/prompt-ed/module-6-scenario-planning-what-ifs
- Type: module | Pillar: S&OP & IBP | Date: 2026-06-24
- Author: Kislay S | Read time: 12 min
- Tags: Learning track, Foundations, Scenario planning, What-if, Overlays
- Summary: The sixth module of the Planning Foundations course. How to model what-if scenarios without disrupting the baseline plan. The overlay approach, comparison, and promotion.
### Module 5: Supply netting and inventory policy
- URL: https://tensoranalytics.ai/prompt-ed/module-5-supply-netting-inventory-policy
- Type: module | Pillar: Inventory & supply | Date: 2026-06-17
- Author: Dhiraj S | Read time: 13 min
- Tags: Learning track, Foundations, Supply netting, Safety stock, Inventory
- Summary: The fifth module of the Planning Foundations course. How to turn the locked demand plan into replenishment orders. Supply netting, safety stock, reorder points, and the gap analysis that prevents stockouts.
### Module 4: The S&OP cycle and sequential gates
- URL: https://tensoranalytics.ai/prompt-ed/module-4-sop-cycle-sequential-gates
- Type: module | Pillar: S&OP & IBP | Date: 2026-06-10
- Author: Kislay S | Read time: 14 min
- Tags: Learning track, Foundations, S&OP, Sequential gates, Consensus
- Summary: The fourth module of the Planning Foundations course. How the S&OP process turns a forecast into a published plan. Sequential gates, consensus, cycle lock, and why most S&OP cycles take three days when they should take four hours.
### Module 3: Forecast accuracy metrics
- URL: https://tensoranalytics.ai/prompt-ed/module-3-forecast-accuracy-metrics
- Type: module | Pillar: Demand forecasting | Date: 2026-06-03
- Author: Dhiraj S | Read time: 13 min
- Tags: Learning track, Foundations, MAPE, WMAPE, Bias
- Summary: The third module of the Planning Foundations course. MAPE, WMAPE, bias, and forecast value added. How to measure accuracy so the number actually reflects reality.
### Module 2: Statistical forecasting models
- URL: https://tensoranalytics.ai/prompt-ed/module-2-statistical-forecasting-models
- Type: module | Pillar: Demand forecasting | Date: 2026-05-27
- Author: Kislay S | Read time: 14 min
- Tags: Learning track, Foundations, Statistical models, ETS, ARIMA
- Summary: The second module of the Planning Foundations course. Exponential smoothing, ARIMA, and seasonal decomposition. What each model actually does, when to use it, and when it will fail.
### Module 1: Foundations of demand planning
- URL: https://tensoranalytics.ai/prompt-ed/module-1-foundations-of-demand-planning
- Type: module | Pillar: Demand forecasting | Date: 2026-05-20
- Author: Kislay S | Read time: 12 min
- Tags: Learning track, Foundations, Demand planning, Course
- Summary: The first module of the Planning Foundations course. What demand planning actually is, who owns it, how it connects to the rest of the business, and the vocabulary you need before the rest of the course makes sense.
### Your ERP data is stale. Here is how to fix it without a multi-year migration.
- URL: https://tensoranalytics.ai/prompt-ed/erp-data-stale-fix-without-migration
- Type: concept | Pillar: Data & integration | Date: 2026-05-13
- Author: Dhiraj S | Read time: 9 min
- Tags: Data quality, ERP, Master data management, Import
- Summary: ERP data quality is the root cause of most planning failures. The fix is not a new ERP. It is a data layer that validates, deduplicates, and versions every record before it reaches your planning tool.
### Human-in-the-loop is not a checkbox. It is a system property.
- URL: https://tensoranalytics.ai/prompt-ed/human-in-the-loop-system-property
- Type: concept | Pillar: AI in operations | Date: 2026-05-06
- Author: Keshab S | Read time: 9 min
- Tags: AI agents, Human-in-loop, Compliance, Audit trail
- Summary: Most vendors claim human-in-loop. Very few implement it. The difference is whether the human has real choice at the decision point, with enough information and time to make a different decision. Here is how to tell the difference.
### Audit trails for AI: what DPDP, CCPA, and the EU AI Act actually require
- URL: https://tensoranalytics.ai/prompt-ed/audit-trails-ai-dpdp-ccpa-eu-ai-act
- Type: concept | Pillar: AI in operations | Date: 2026-04-29
- Author: Keshab S | Read time: 12 min
- Tags: Compliance, Audit trail, DPDP Act, CCPA, EU AI Act
- Summary: Three regulatory frameworks now require audit trails for automated decisions. The requirements overlap but are not identical. Here is what each requires, where they differ, and what to implement if you operate in all three jurisdictions.
### The safety stock formula is not wrong. It is misused.
- URL: https://tensoranalytics.ai/prompt-ed/safety-stock-formula-misused
- Type: concept | Pillar: Inventory & supply | Date: 2026-04-22
- Author: Dhiraj S | Read time: 10 min
- Tags: Inventory, Safety stock, Reorder points, Service level
- Summary: Every supply chain textbook prints the safety stock formula. Most practitioners apply it without checking whether its assumptions hold for their SKUs. When the assumptions fail, the formula produces confident nonsense.
### Why scenario planning fails: the overlay problem
- URL: https://tensoranalytics.ai/prompt-ed/scenario-planning-overlay-problem
- Type: concept | Pillar: S&OP & IBP | Date: 2026-04-15
- Author: Kislay S | Read time: 7 min
- Tags: S&OP, Scenario planning, What-if analysis, Planning
- Summary: Most scenario planning tools copy the entire baseline plan and let you edit the copy. This breaks the moment the baseline changes. The fix is sparse overlays, and almost no tool does it correctly.
### The 4-hour S&OP cycle is not a goal. It is a consequence of design.
- URL: https://tensoranalytics.ai/prompt-ed/four-hour-sop-cycle
- Type: concept | Pillar: S&OP & IBP | Date: 2026-04-08
- Author: Kislay S | Read time: 8 min
- Tags: S&OP, Process design, Cycle time, Consensus
- Summary: Most S&OP cycles take three days because they are designed to take three days. The cycle time is a function of process architecture, not effort. Change the architecture and the time collapses.
### The M5 competition settled the ensemble question. Most planners haven't noticed.
- URL: https://tensoranalytics.ai/prompt-ed/m5-competition-ensemble-question
- Type: concept | Pillar: Demand forecasting | Date: 2026-04-02
- Author: Dhiraj S | Read time: 10 min
- Tags: Forecasting, Ensemble, M5 Competition, Machine Learning, N-BEATS
- Summary: The M5 forecasting competition was the largest controlled study of forecasting methods ever run. The results, published in 2022, are clear. Ensembles win. Here is what they showed, and why most planning tools still ship single-model forecasting.
### Forecasting intermittent demand: what Croston got right, and what came after
- URL: https://tensoranalytics.ai/prompt-ed/forecasting-intermittent-demand-croston
- Type: concept | Pillar: Demand forecasting | Date: 2026-03-25
- Author: Dhiraj S | Read time: 11 min
- Tags: Forecasting, Intermittent demand, Croston, SBA, Statistical models
- Summary: Croston's method has been the default for intermittent demand since 1972. It is still useful. But three decades of research have produced better alternatives. Here is what to use when.
### What MAPE actually tells you (and what it hides)
- URL: https://tensoranalytics.ai/prompt-ed/what-mape-actually-tells-you
- Type: concept | Pillar: Demand forecasting | Date: 2026-03-18
- Author: Dhiraj S | Read time: 9 min
- Tags: Forecasting, MAPE, Accuracy metrics, Demand Planning
- Summary: MAPE is the most quoted forecast accuracy metric in supply chain. It is also the most misread. Here is what the number actually means, what it conceals, and what to track alongside it.
### Why One Forecast Model Is Never Enough
- URL: https://tensoranalytics.ai/prompt-ed/why-one-forecast-model-is-never-enough
- Type: concept | Pillar: Demand forecasting | Date: 2026-03-10
- Author: Kislay S | Read time: 8 min
- Tags: Forecasting, Technical, Ensemble, Demand Planning
- Summary: Every SKU has a different demand fingerprint. A single model applied across the range hides the signal. The answer is not a better model. It is a blend of models, picked per series.
### AI Regulation in Industry 5.0: Why Ethical AI Is Harder Than It Sounds
- URL: https://tensoranalytics.ai/prompt-ed/ai-regulation-in-industry-5
- Type: editorial | Pillar: The AI Hype Hangover | Date: 2026-02-13
- Author: Ayushi P | Read time: 10 min
- Tags: AI Hype Hangover, Editorial, Regulation, Ethics, Industry 5.0
- Summary: There's a widening gap between what the tech world promises and what's actually being built on factory floors. Why ethical AI still means looking past the glossy presentations at industry conferences.
### When the Emperor's Code Has No Clothes
- URL: https://tensoranalytics.ai/prompt-ed/when-the-emperors-code-has-no-clothes
- Type: editorial | Pillar: The AI Hype Hangover | Date: 2025-12-24
- Author: Kislay S | Read time: 7 min
- Tags: AI Hype Hangover, Editorial, Enterprise AI, ROI
- Summary: Discover why most enterprise AI projects are struggling to deliver ROI in 2025. From high-profile layoffs to the funding bubble, here's the harsh truth about GenAI ROI and where the actual value sits.
## Frequently asked questions

### What is GrepEye?
GrepEye is Tensor Analytics' operational intelligence platform with eight pillars: data infrastructure, rules and automation, analytics, integrated business planning, AI agent orchestration, forms, access control, and config packs.

### Who is Tensor Analytics for?
Mid-market manufacturers and distributors ($50M-$500M revenue) running S&OP in Excel and ERP. Personas include demand planners, S&OP leads, supply chain managers, and heads of operations.

### How do I book a demo?
Request a walkthrough at https://tensoranalytics.ai/contact or email teams@tensoranalytics.ai.

## Contact

- Sales and pilots: teams@tensoranalytics.ai
- General enquiries: info@tensoranalytics.ai
- Entity: Tensor Analytics LLC