MatSimKit-Design

A ChatGPT-like interface lets engineers describe calculation needs in everyday language — for example, "Calculate the solidification range of IN718" or "Compare the high-temperature strength of two stainless steels" — and the AI Agent automatically interprets the intent, calls the JMatPro® engine, and presents results as charts and professional recommendations, with no knowledge of software operation required. Multi-turn dialogue with full history management allows engineers to continue questioning and deepen analysis within the same session.

The software covers all nine alloy types — iron alloys, cobalt alloys, nickel alloys, aluminum alloys, magnesium alloys, titanium alloys, copper alloys, zirconium alloys, and tin alloys — and supports thermodynamic calculations, thermophysical property calculations, solidification calculations, mechanical property calculations, kinetic phase transformation calculations, and more. From stating a requirement to receiving a chart typically takes just a few minutes, drastically compressing traditional calculation workflows.

Users can upload experimental measurement data in CSV format; the AI automatically parses the content and overlays it with calculation results for direct comparison, clearly showing discrepancies between computed and measured values. This helps engineers assess model accuracy and pinpoint the root cause of performance deviations, enabling a calculation-plus-experiment dual-driven R&D workflow.

The sensitivity analysis function lets users scan a range of element content; the system batch-calculates how that element's variation affects target properties and plots the influence curve, quantifying each element's contribution weight. This provides data-backed guidance for composition fine-tuning and process parameter optimization, effectively replacing large numbers of single-factor experiments.