The field of failure modes and effects analysis (FMEA) has been a cornerstone of engineering and quality management for decades, allowing organizations to proactively identify and mitigate potential failures before they occur. Traditional FMEA methods are often time-consuming and labor-intensive, requiring significant expertise and knowledge to be effective. Recent advancements in artificial intelligence (AI) and machine learning (ML) have opened new possibilities for FMEA, allowing organizations to harness the power of language models like ChatGPT to transform their FMEA process and unlock new insights [1].

ChatGPT is a language model developed by OpenAI, which uses deep learning algorithms to generate human-like responses to text-based inputs. ChatGPT can understand the context of a given text, generating appropriate responses, and learning from new data to continuously improve its performance. The information presented has been gathered from prior FMEA investigations that were previously conducted. It is derived from a compilation of more than 100 trillion past data sources, representing the amassed knowledge acquired from online records. This technology has significant potential in FMEA, enabling organizations to generate and evaluate failure modes, identify potential consequences, and recommend mitigation strategies more quickly and accurately than ever before.

GPT stands for “Generative Pre-trained Transformer.” It is a type of deep learning model that uses a transformer architecture to generate human-like language. The model is pre-trained on large amounts of text data, allowing it to generate coherent and contextually appropriate responses to prompts or questions [2]. A schematic representation of the ChatGPT workflow is shown in Fig. 1. The transformer consists of multiple layers of nodes that process information from both past and future inputs, allowing it to understand the context of the text it is generating.

Fig. 1
figure 1

The system takes input data, such as historical failure training data, design information, and engineering knowledge, and uses deep learning algorithms to generate and evaluate failure modes, identify potential consequences, and recommend mitigation strategies. The model is continuously trained and refined over time, allowing for even greater accuracy and effectiveness in FMEA processes

Full size image

GPT is pre-trained on massive amounts of text data using unsupervised learning techniques, allowing it to learn patterns and relationships in language. Once pre-trained, it can be fine-tuned on specific tasks such as text completion, translation, summarization, and question–answering.

When generating text, GPT takes a prompt as input and generates a continuation based on the patterns and relationships it learned during pre-training. GPT generates text one word at a time, with each word being conditioned on the previous words in the sequence. The final output is a coherent piece of text that is like human-generated language.

The training data for GPT comes from large corpora of text, such as books, articles, and websites. The data is typically sourced from the Internet or curated by organizations like OpenAI, who trained GPT on a diverse range of text, including books, articles, and websites.

In terms of reinforcement data, GPT can be fine-tuned on specific tasks using supervised learning techniques. For example, to generate text that is more relevant to a specific domain, GPT can be fine-tuned on a smaller dataset of text from that domain. This fine-tuning process provides the reinforcement data that helps GPT generate more accurate and relevant text for a particular task.

The first step was to train the ChatGPT model on the existing FMEA data, allowing the model to learn from the manufacturer’s previous experiences and generate more accurate and relevant results. The manufacturer was then able to use ChatGPT to quickly generate and evaluate failure modes for various components and systems, identifying potential consequences and recommending mitigation strategies. This process significantly reduced the time and cost of the FMEA process while also increasing the accuracy and thoroughness of the results.

The power of ChatGPT in FMEA lies in its ability to generate and evaluate failure modes, identify potential consequences, and recommend mitigation strategies quickly and accurately. Here is how it works:

  1. 1.

    Training the model The first step in using ChatGPT for FMEA is to train the model on existing FMEA data. This allows the model to learn from previous experiences and generate more accurate and relevant results.

  2. 2.

    Generating failure modes Once the model is trained, it is used to generate a list of potential failure modes for a given system or process. ChatGPT uses deep learning algorithms to analyze previous data and knowledge to generate these failure modes, ensuring that the possible failure modes are considered.

  3. 3.

    Identifying potential consequences After generating failure modes, ChatGPT can be used to identify potential consequences of each failure mode. Again, the model uses deep learning algorithms to analyze previous data and knowledge to ensure that the potential consequences of a failure mode are identified.

  4. 4.

    Assessing risk ChatGPT can also be used to help assess the risk associated with each failure mode and its potential consequences. By analyzing the likelihood and severity of each failure mode and consequence, ChatGPT can help prioritize mitigation strategies based on the greatest potential impact on safety, quality, and cost.

  5. 5.

    Recommending mitigation strategies Finally, ChatGPT can recommend mitigation strategies based on the identified failure modes and potential consequences. These strategies may include design changes, process improvements, or other actions that can reduce the likelihood or severity of the failure modes and their consequences.

The potential of ChatGPT in FMEA is just beginning to be realized, and the future looks bright. As more organizations adopt AI and ML technologies, we can expect to see even greater improvements in the efficiency, accuracy, and effectiveness of FMEA processes. One of the key benefits of ChatGPT is its ability to continuously learn and improve over time. As the model is exposed to new data and experiences, it can adapt and refine its understanding of failure modes and their consequences, allowing for even more accurate and insightful results. In addition, ChatGPT has the potential to be used in conjunction with other AI and ML technologies, such as computer vision and natural language processing, to create even more powerful and comprehensive FMEA systems. By combining these technologies, organizations can gain deeper insights into potential failure modes and their consequences, leading to more effective mitigation strategies and greater levels of safety and quality.

One example of the transformative potential of ChatGPT in FMEA is in the aerospace manufacturing industry. A large aerospace manufacturer was facing increasing pressure to reduce the time and cost of their FMEA process while still maintaining high levels of quality and safety. The manufacturer decided to partner with a technology provider to implement ChatGPT in their FMEA process.

As AI and ML technologies continue to advance, we can expect to see even greater improvements in the efficiency and effectiveness of FMEA processes. ChatGPT is just the beginning of what promises to be an exciting future for FMEA and the organizations that use it to ensure safety, quality, and reliability.