// Decision Engine 06

Full Deployment
Planner

Define your complete deployment scenario across all dimensions. The engine calculates hardware, power infrastructure, network bandwidth, and storage endurance — and outputs a full bill of materials with cost estimates.

// Define deployment

01
Deployment Scenario
// Primary use case
Retail Analytics
Industrial Inspection
Smart City / Traffic
Security / Surveillance
Robotics / Autonomous
Agriculture / Remote
02
Camera Count
// Total concurrent video streams
1
2
4
8
16
32
// Drives compute, network, and storage requirements
03
Camera Resolution
// Per-stream resolution
720p HD
1080p FHD
4K UHD
04
AI Model Type
// Primary inference workload
Classification
Object Detection
Segmentation
Pose Estimation
Multi-Model Pipeline
05
Power Infrastructure
// Site power delivery method
Mains / AC Power
PoE Network Switch
Battery / Solar
UPS Backed Mains
06
Video Retention
// Local storage retention window
1d
7d
30d
90d
365d
// Affects storage capacity and drive endurance requirements
07
Deployment Environment
// Physical installation context
Indoor / Controlled
Indoor / Industrial
Outdoor / Sheltered
Outdoor / Exposed
Mobile / Vehicle
// Select all parameters to continue
// Computing full deployment specification…

// Bill of Materials

// Deployment Specification
SPEC COMPLETENESS
// machine-readable BOM — application/json

        

What this Full Deployment Planner decides

This planner is the deployment-level decision layer for EdgeAIStack. It combines compute selection, power infrastructure, network sizing, storage retention, enclosure requirements, and installation cost into a single deployment specification. The goal is to turn a scenario description into a practical, machine-readable bill of materials for real edge AI deployments.

// Inputs considered
01
Scenario + Model

Use case and AI model type determine the practical compute profile and deployment shape.

02
Cameras + Resolution

Stream count and resolution drive compute demand, network bandwidth, and storage growth.

03
Power + Retention + Environment

Power method, storage retention, and deployment environment shape the infrastructure bill of materials and total project cost.

// What the planner assembles

The planner takes the selected deployment inputs and generates a full-system estimate across the major infrastructure layers needed for edge AI: compute hardware, network switching, power delivery, storage media, enclosures, and installation. The result is designed to help engineers and buyers move from idea to deployment specification faster.

  • Compute platform recommendation and quantity
  • Network switch and cabling estimate
  • Power infrastructure class and load planning
  • Storage capacity and endurance fit
  • Environmental enclosure guidance
  • Installation and project-level cost estimate
// How the bill of materials is structured

The output groups costs and components into deployment sections so the result is readable by both humans and downstream systems. It is intended as a planning BOM rather than a final quote.

  • Summary layer: project cost, hardware cost, power draw, network bandwidth, storage requirement, and selected accelerator
  • BOM sections: compute, network, power, storage, enclosure, and installation
  • Warnings layer: constraints such as tight compute headroom, storage fit risk, power limits, or bandwidth bottlenecks
  • Machine-readable layer: exportable JSON for configuration reuse, sharing, or API-based workflows
// Worked examples
// Example 01
Retail analytics starter deployment
A small 1–4 camera indoor deployment with 1080p streams and classification or detection usually lands on a lightweight compute platform with modest power and storage requirements.
// Example 02
Industrial or security multi-camera system
An 8–16 camera 1080p or 4K deployment typically increases compute class, switch capacity, and retention storage, while enclosure and power delivery become more important.
// Example 03
Outdoor or mobile rugged deployment
Outdoor, exposed, or mobile installations shift the BOM toward ruggedized enclosures, more careful power design, and deployment-specific infrastructure tradeoffs.
// Example machine-readable output

      
// application/json full-deployment-planner/v1
{
  "tool": "full-deployment-planner",
  "schema_version": "v1",
  "inputs": {
    "scenario": "security",
    "cameras": 8,
    "resolution": "1080p",
    "model": "detection",
    "power_infra": "poe",
    "retention": 30,
    "environment": "outdoor_sheltered"
  },
  "outputs": {
    "summary": {
      "total_project_cost": 4250,
      "hardware_cost": 3530,
      "total_power_w": 140,
      "network_mbps": 48,
      "storage_tb": 15.4,
      "accelerator": "Jetson Orin Nano"
    },
    "bill_of_materials": [
      {
        "item": "Jetson Orin Nano",
        "qty": 2,
        "est_cost_usd": 998
      },
      {
        "item": "Managed PoE Switch",
        "qty": 1,
        "est_cost_usd": 220
      },
      {
        "item": "NVMe Storage",
        "qty": 2,
        "est_cost_usd": 320
      }
    ]
  }
}

// FAQ

Is this the main planning tool on the site?

Yes. This is the deployment-level tool that brings together the reasoning from the individual hardware, power, network, and storage calculators into one consolidated planning output.

Does the bill of materials include installation?

Yes. The planner includes an installation estimate as part of the deployment-level total, though it should be treated as a planning-grade estimate rather than a final services quote.

Can the output be shared or reused?

Yes. The planner produces machine-readable JSON output and supports shareable configurations so the deployment can be reviewed, exported, or reused in downstream workflows.

Should I send the second file too?

If the slug has an additional file that renders saved configurations, print views, or deployment-specific output pages, that file is worth reviewing too because it likely affects crawlability and how AI systems interpret saved deployment specs.