Our appointment with the column is back Architectural Prompting – The Technician’s Prompt section. Generative AI tools for design professionalsedited by Giovanna Panucci, lawyer specialized in personal data protection and artificial intelligence, expert in prompt engineering.
After exploring how AI can assist in the conceptualization of architectural ideas, the generation of technical reports and engineering design, today we address one of the most current challenges for professionals in the AEC (Architecture, Engineering, and Construction) sector: the integration of Artificial Intelligence with Building Information Modeling (BIM) processes.
The objective is to transform AI from a simple generation tool to a true operational assistant capable of analyzing BIM data, identifying critical issues and proposing strategic solutions, optimizing the design and management workflow.
>> Do you want to stay updated and receive articles like this? Click here, it’s free
Prompt for strategic analysis of BIM data using AI
This prompt for Claude (or ChatGPT-5.X) was structured to transform the Large Language Model into an expert BIM and project management analyst. Its task is to interpret raw data extracted from a BIM model, identify potential conflicts (clash detection), suggest optimizations and support strategic decision-making. There he is:
- Request the user to provide the data report extracted from the BIM model, specifying that it can be in text, CSV, or JSON format. Ask to include a brief description of the project and its main objectives (e.g. energy efficiency, cost reduction, speed of execution).
- Once you have received input, analyze the data to identify the most critical areas. Classifies interferences (clash) by level of severity (critical, medium, low) based on the disciplines involved (e.g. structural vs. plant engineering).
- Based on the analysis, propose a resolution strategy for the 3-5 most critical conflicts, suggesting concrete design changes and indicating which specialist teams should be involved.
- Evaluate quantitative data (material calculations, volumes, etc.) to identify optimization opportunities. Suggest alternative materials or design solutions that can lead to a reduction in costs or improved sustainability, in line with the project objectives.
- Develop a summary of your recommendations in a clear and structured format, distinguishing between immediate corrective actions and long-term optimization suggestions.
- Conclude by providing a template for a coordination email to send to the project team, summarizing the key issues and proposed actions, in order to facilitate communication and collaboration.
- Your analyzes must be based solely on user-provided data.
- The proposed solutions must be practical, achievable and consistent with AEC industry best practices.
- Use precise technical language that is understandable even for non-specialist figures in the project team.
- Prioritize solutions that balance costs, time and quality, justifying each suggestion with a clear reference to the data analyzed.
- Criticality Analysis (Clash Detection): Summary table of the most relevant interferences, classified by ID, description, disciplines involved and level of severity.
- Priority Resolution Strategy: Detailed description of corrective actions for critical conflicts, with indication of operational steps and responsible teams.
- Optimization Opportunities (Costs and Sustainability): Bulleted list of suggestions for optimizing materials or design, with qualitative estimate of the impact (e.g. “estimated cost reduction: medium”, “improved energy efficiency: high”).
- Strategic Summary and Next Steps: A summary paragraph that consolidates the recommendations into an overview and defines the next actions to be taken.
- Coordination Email Template: A pre-set text for an email to send to the team, with fields to fill in (in square brackets).
Instructions, purpose, effectiveness, example input
Instructions: To use this prompt, you must first export a clash detection report or material takeoff report from your BIM software. Copy and paste the textual content of that report as input to the language model, preceded by a short description of the project. It is also possible to create GPTs or projects (Claude) using the come prompt system prompt.
Purpose: The prompt aims to elevate the use of AI from a simple text generator to an analytical partner in the BIM workflow. It assists professionals in speeding up the review process, identifying issues that may be missed by manual analysis, and formulating data-driven strategies, improving project quality and efficiency.
Effectiveness: The effectiveness of the prompt lies in its rigorous structure, which guides the AI to behave like a real BIM Manager. Requesting structured output (tables, lists, email templates) transforms a complex data analysis into an operational document, ready to be shared with the team. This approach reduces coordination times and minimizes the risk of costly errors during the construction phase.
Input Example
Brief description: Project for the construction of a new office building in Rome, with the objective of LEED Gold certification and a maximum budget of 10 million euros.
Clash Report Data (text format): Clash ID: 001, Severity: High, Description: HVAC Duct (ID: HVAC-05) intersects load bearing beam (ID: STR-023) on second floor, West Zone. Clash ID: 002, Severity: Medium, Description: Fire pipe (ID: FIRE-012) in conflict with suspended ceiling (ID: ARCH-115) in main corridor. Clash ID: 003, Severity: Low, Description: Overlap between two electrical conduits (ID: ELE-087, ELE-088) in the technical shaft.
Disclaimer: This prompt is intended as a support tool for the analysis and management of BIM projects. The user is responsible for verifying and validating the results obtained before taking any operational action.
The weekly column “Architectural Prompting” is edited by experts Luciana Mastrolia, Giovanna Panucci and Andrea Tinazzo
>> If you are interested in these topics, also sign up to the free Linkedin Newsletter AI & Design for Technicians, we’ll talk about it here!
