AI Equipment Recommendations

NexusRMS includes an AI-powered equipment suggestion engine that analyses your historical booking data to recommend relevant equipment when you are building projects. Instead of manually searching for items you might need, the system surfaces intelligent suggestions based on patterns it has learned from your completed projects.

How it works

The recommendation engine runs in the background, continuously learning from your completed projects. It examines which equipment items are booked together, which clients tend to hire specific items, and which venues require particular gear. When you add equipment to a new project, the engine compares your current selections against these learned patterns and surfaces suggestions ranked by confidence.

Where recommendations appear

Recommendations appear in the project equipment selection workflow. When you are adding equipment to a project, a Suggestions panel appears alongside the equipment browser. Each suggestion shows the equipment name, image thumbnail, a confidence score, and a one-click Add to Project button. Suggestions update in real time as you add or remove items from the project.

Recommendation types

The engine produces three types of recommendations, each drawing from a different data source:

Client History

Client History recommendations suggest equipment that the same client has frequently booked in previous projects. For example, if a client has hired a specific lighting rig on their last five bookings, the system suggests that rig when you create a new project for the same client. This helps you quickly replicate a client's preferred setup.

Venue Patterns

Venue Patterns recommendations suggest equipment commonly used at the specific venue assigned to the project. Different venues have different requirements — a particular conference hall might always need a specific PA configuration, while an outdoor festival site might always require weatherproof covers and long cable runs. The engine learns these venue-specific patterns from historical data.

Co-Usage Patterns

Co-Usage Patterns recommendations suggest items that are frequently booked alongside items already in your project. For example, if you add a mixing desk, the engine might suggest XLR cables, DI boxes, and monitor speakers because those items have historically been booked together. This is the most powerful recommendation type and helps ensure you do not forget essential accessories or companion equipment.

Confidence scoring

Each recommendation carries a confidence score from 0% to 100%, calculated based on the strength of the underlying data pattern. A score of 90% means the suggested item was booked in 90% of similar historical situations. Higher scores indicate stronger, more reliable patterns. The confidence score is displayed as a percentage badge on each suggestion card.

Configuration

All AI recommendation settings are managed in Configuration > Equipment > AI Suggestions tab. The following options are available:

• Enable AI Suggestions — Master toggle to turn recommendations on or off. Default: On.
• Minimum Confidence Threshold — Only suggestions at or above this percentage are shown. Range: 0–100%. Default: 50%. Lower values show more suggestions but with less certainty; higher values show fewer but more reliable suggestions.
• Maximum Suggestions — The maximum number of suggestions displayed at once. Range: 1–20. Default: 5.
• Recommendation Types — Toggle each of the three recommendation types (Client History, Venue Patterns, Co-Usage Patterns) independently. All three are enabled by default.
• Show Availability Status — When enabled, each suggestion displays real-time availability for the project's date range, so you can see immediately whether the suggested item is free. Default: On.

Auto-learning

The recommendation engine automatically improves over time. Every time a project is marked as completed, the engine incorporates that project's data into its pattern library. You do not need to train the system manually — it learns passively from your normal workflow. The more projects you complete, the more accurate and relevant the suggestions become.

Data requirements

The engine needs a minimum of 20 completed projects before it can generate meaningful suggestions. Until this threshold is reached, the Suggestions panel displays a message explaining that the system is still learning. The quality and quantity of suggestions improve steadily as you complete more projects beyond this baseline.

Tips

• Start with a 50% confidence threshold — The default threshold provides a good balance between suggestion quantity and quality. Adjust downward if you are seeing too few suggestions, or upward if the suggestions feel too broad.
• Enable all three recommendation types — Each type covers a different angle. Using all three together gives you the most comprehensive suggestions.
• The system gets smarter with every completed project — Make sure to mark projects as completed when they finish. This feeds new data into the engine and keeps suggestions current.
• Review suggestions rather than blindly accepting them — AI suggestions are recommendations, not requirements. Always review each suggestion against the specific needs of the current project before adding it.
• Use venue patterns for recurring venue clients — If you regularly service the same venues, venue-based suggestions save significant time by pre-populating the gear list based on what has worked before.

Next steps

Continue to the next article for troubleshooting common equipment issues, including problems with scanning, stock levels, pricing, and serial number generation.