BMS-AI-BOT / BMS_Tech_Specs_Pro.html
BMS Deployer
Remove header and fix inventory
9916f7c
raw
history blame
5.47 kB
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>BMS AI Assistant - Technical Specifications</title>
<style>
:root {
--primary: #D01010;
--header-bg: #333;
--light-bg: #f8f9fa;
}
body {
font-family: 'Segoe UI', Arial, sans-serif;
line-height: 1.6;
color: #333;
max-width: 1000px;
margin: 0 auto;
padding: 40px;
}
header {
border-bottom: 4px solid var(--primary);
margin-bottom: 40px;
padding-bottom: 20px;
}
h1 {
margin: 0;
color: var(--header-bg);
font-size: 2.5em;
}
h2 {
color: var(--primary);
margin-top: 40px;
border-bottom: 1px solid #ddd;
padding-bottom: 10px;
}
h3 {
color: #555;
margin-top: 25px;
}
table {
width: 100%;
border-collapse: collapse;
margin: 20px 0;
font-size: 0.95em;
}
th,
td {
padding: 12px 15px;
border: 1px solid #ddd;
text-align: left;
}
th {
background-color: var(--header-bg);
color: white;
font-weight: 600;
}
tr:nth-child(even) {
background-color: var(--light-bg);
}
.tech-grid {
display: grid;
grid-template-columns: repeat(2, 1fr);
gap: 20px;
}
.tech-box {
background: var(--light-bg);
padding: 20px;
border-left: 4px solid var(--primary);
}
.tech-box h4 {
margin-top: 0;
color: var(--primary);
}
ul {
margin: 0;
padding-left: 20px;
}
</style>
</head>
<body>
<header>
<h1>BMS AI Assistant</h1>
<p><strong>System Technical Specifications</strong> | Version 1.0</p>
</header>
<h2>1. System Overview</h2>
<p>The BMS AI Assistant is a specialized conversational interface designed for supply chain optimization. It
integrates deterministic forecasting models with natural language processing to provide real-time decision
support for inventory and procurement operations.</p>
<h2>2. Technology Stack</h2>
<div class="tech-grid">
<div class="tech-box">
<h4>Core Framework</h4>
<ul>
<li><strong>Language:</strong> Python 3.10</li>
<li><strong>API Framework:</strong> FastAPI</li>
<li><strong>Server:</strong> Uvicorn (ASGI)</li>
</ul>
</div>
<div class="tech-box">
<h4>Data & Analytics</h4>
<ul>
<li><strong>Processing:</strong> Pandas, NumPy</li>
<li><strong>Forecasting:</strong> Statsmodels (ARIMA)</li>
<li><strong>Storage:</strong> Flat-file CSV (In-memory cache)</li>
</ul>
</div>
<div class="tech-box">
<h4>Frontend</h4>
<ul>
<li><strong>Structure:</strong> HTML5</li>
<li><strong>Styling:</strong> CSS3 (Custom Design System)</li>
<li><strong>Logic:</strong> Vanilla JavaScript (ES6+)</li>
</ul>
</div>
<div class="tech-box">
<h4>Deployment</h4>
<ul>
<li><strong>Containerization:</strong> Docker</li>
<li><strong>Platform:</strong> Hugging Face Spaces</li>
<li><strong>CI/CD:</strong> Git-based Workflow</li>
</ul>
</div>
</div>
<h2>3. Component Specifications</h2>
<table>
<thead>
<tr>
<th>Component</th>
<th>Specification</th>
<th>Function</th>
</tr>
</thead>
<tbody>
<tr>
<td><strong>Forecasting Engine</strong></td>
<td>ARIMA (1,1,1) Model</td>
<td>Generates 30-day demand projections based on historical time-series data.</td>
</tr>
<tr>
<td><strong>Intent Parser</strong></td>
<td>Regex / Pattern Matching</td>
<td>Identifies user intent (Forecast, Inventory, Supplier) with <50ms latency.</td>
</tr>
<tr>
<td><strong>LLM Integration</strong></td>
<td>TinyLlama 1.1B (GGUF)</td>
<td>Handles unstructured general queries and conversational context.</td>
</tr>
<tr>
<td><strong>PDF Generator</strong></td>
<td>FPDF Library</td>
<td>Dynamically compiles chat transcripts and data tables into downloadable reports.</td>
</tr>
</tbody>
</table>
<h2>4. Performance Metrics</h2>
<ul>
<li><strong>API Latency:</strong> &lt; 100ms (Standard Queries), &lt; 2s (Forecasting)</li>
<li><strong>Concurrent Sessions:</strong> Scalable via Docker replicas (Default: 2 vCPUs)</li>
<li><strong>Uptime:</strong> 99.9% (Hugging Face Infrastructure)</li>
</ul>
</body>
</html>