Tera-Calculation EIQ_AI Integrated Portal

AI Logistics Model Generation | Next-Gen EIQ Analysis

Switch Language within AI Portal

Mission Statement

This software is a logistics learning tool. It is designed to assist logistics professionals and students in learning how to create logistics AI software.
Currently, the prototype is being used by two universities, and the final version is planned to be released as free software.
Additionally, this is the AI version of our "Tera-Calculation" software, which teaches specific EIQ analysis calculation methods.

Next-Gen Logistics Analysis: EIQ × Machine Learning

This project combines 30 years of consulting expertise with the latest machine learning (Microsoft.ML) to instantly derive optimal logistics models from actual shipping data. High-precision logistics models are automatically generated from minimal information: Number of Destinations (E) and Items (I).

1. Training & Construction

Train SDCA regression models using any shipping data. Generates AI intelligence capable of predicting over 160 analysis items.

Details: EIQ_AI Model Specs →
2. Prediction & Matrix Visualization

Using the generated model, instantly output and visualize 6 different matrices, including frequency, weight, and volume.

Details: Software Workflow →

Future Roadmap

  • Inventory & Space Integration: Integrate maximum/safety stock conditions into predictive matrices. Automatically calculate required stock, incoming volumes, and facility layout areas.
  • "Tera-Calculation" Series Integration: Enable rapid facility evaluation as a preliminary analysis tool within the series.
  • Pursuit of Precision: Aim for prediction error rates within 3% through continuous model refinement.
📌 Planned Updates for Official Release:
The official release will separate "Piece" and "Case" shipping predictions. We will also add Quantity (Q) and Lines (R) as parameters to evolve into a universal model for fractal analysis across all industries and scales.

Technology: Microsoft.ML (ML.NET)

This portal utilizes "ML.NET," an open-source framework that allows building custom machine learning models within a .NET environment. It enables advanced logistics analysis purely within C#, without depending on external languages like Python.

1. ML.NET Core
  • Cross-Platform: Runs on Windows, Linux, and macOS.
  • High Performance: Optimized .NET 6.0+ for high-speed processing.
2. SDCA Regression
  • Efficient Algorithm: Fast learning with a low memory footprint.
  • Optimized for Logistics: High-precision weight/volume prediction based on E and I factors.