About Prompt
- Prompt Type – Dynamic
- Prompt Platform – ChatGPT, Grok, Deepseek, Gemini, Copilot, Midjourney, Meta AI and more
- Niche – Defect Risk Assessment
- Language – English
- Category – Predictive Quality Analytics
- Prompt Title – AI Prompt for Predicting Product Failures Based on Manufacturing Data
Prompt Details
**Prompt Type:** Dynamic
**Niche:** Defect Risk Assessment for Predictive Quality Analytics
**Target AI Platforms:** All
**Prompt Objective:** Predict the probability of product failure and identify contributing factors based on provided manufacturing data. The prompt should be adaptable to various data structures and manufacturing processes.
**Prompt Structure:**
This prompt utilizes a dynamic structure, allowing users to specify the product, data format, and desired output. It incorporates best practices like clear instructions, explicit data formatting expectations, and output specifications for improved performance.
“`
## Product Failure Prediction Prompt
**1. Product Definition:**
* **Product Name:** [Enter the name of the product, e.g., “Circuit Board X7”]
* **Product Description:** [Provide a concise description of the product and its intended function, e.g., “A high-density circuit board used in automotive control systems.”]
* **Key Failure Modes:** [List the known or suspected failure modes for this product, e.g., “Short circuit, component failure, solder joint fatigue.”]
**2. Manufacturing Data Input:**
* **Data Format:** [Specify the format of the input data. Options: CSV, JSON, XML, SQL Database. If other, please specify.]
* **Data Description:** [Provide a detailed description of the data, including column names and their meaning. For example:
* “timestamp”: Date and time of data collection.
* “machine_id”: Identifier for the manufacturing machine.
* “temperature”: Operating temperature during manufacturing.
* “pressure”: Pressure applied during a specific process.
* “component_id”: Identifier for individual components used in the product.
* “test_result_1”: Result of the first quality control test.
* “test_result_2”: Result of the second quality control test. (…and so on)]
* **Data Location:** [Specify where the data is located. Options: Upload file, URL, Database connection string. If other, please specify.]
**3. Prediction Task:**
* **Prediction Target:** [Specify what you want to predict. Options:
* Probability of failure within a specific timeframe (e.g., “Probability of failure within the first year of operation”).
* Specific failure mode (e.g., “Predict the most likely failure mode”).
* Remaining Useful Life (RUL) (e.g., “Predict the remaining useful life of the product in hours”).]
* **Time Horizon (if applicable):** [Specify the time horizon for the prediction, e.g., “1 year”, “10,000 operating hours”.]
**4. Output Requirements:**
* **Output Format:** [Specify the desired output format. Options: Table, JSON, Text summary. If other, please specify.]
* **Desired Metrics:** [Specify the desired evaluation metrics. Examples: Accuracy, Precision, Recall, F1-score, AUC-ROC.]
* **Explanation (Optional):** [Specify if you require an explanation for the predictions, e.g., “Identify the top 3 factors contributing to the predicted failure probability.”]
**Example Data Snippet (Optional):** [Include a small sample of the data to demonstrate the structure and format. This can significantly improve the model’s understanding of the input.]
**Specific Instructions (Optional):** [Provide any specific instructions or constraints, e.g., “Consider the correlation between temperature and pressure during manufacturing,” “Focus on failures related to component_id ‘A123’.”]
**Prompt Conclusion:**
Based on the provided information, generate a prediction for the specified target, adhering to the output requirements. Ensure the predictions are well-justified and, if requested, include explanations for the results.
“`
**Using the Prompt:**
1. **Fill in the placeholders:** Replace the bracketed information with the specific details for your product and manufacturing data.
2. **Adapt the prompt:** Modify sections as needed to align with your particular use case and data characteristics. For instance, add or remove fields, adjust the data description, and refine the prediction task.
3. **Iterate and refine:** Experiment with different prompt variations and analyze the outputs. Refine the prompt based on the results to improve the accuracy and relevance of the predictions.
This dynamic prompt allows flexibility and customization for a variety of product failure prediction tasks using diverse manufacturing data. By clearly defining the product, data, prediction task, and output requirements, you can leverage the power of AI to improve product quality and reduce failure rates.