Holistic approach to enterprise artificial intelligence adoption AI technology can significantly reduce the time and effort of many kinds of experts required to carry out HAZOP studies
Aridaman Singh Ahir, Rohit Dombi, and Harirajan Padmanabhan hpad, Inc.
E very enterprise deals with complex, multidisciplinary problems with many internal and external dependencies. Enterprises rely on subject matter experts (SMEs) to solve such complex problems by limiting the scope of the problem and analysing the problem space collaboratively. While collaborating, SMEs apply their knowledge to create solutions. These solutions are, in fact, artefacts that result from applying this knowledge. The knowledge itself is not incorporated into the solutions. Solutions developed in this manner through human knowledge and analysis rely on the mental models of SMEs. Such solutions are invariably fragmented and siloed in their nature because the mental models of SMEs are fragmented, and their knowledge is limited to their domain of expertise. Furthermore, when solution requirements or conditions in their context change, the solutions developed in this manner will need further SME collaboration to analyse the changes and update the solution. This gives rise to numerous complexities and inefficiencies within an enterprise, resulting in the underperformance of the whole enterprise. Therefore, there is a need for a more holistic approach to solve enterprise problems. hpad, Inc. (hpad) has developed technology to address this need for a more holistic approach. This technology involves using AI to gather the collective knowledge of the enterprise, including the mental models of SMEs, and encode this knowledge into an enterprise AI model. In this technology, knowledge from SMEs is gathered and incorporated iteratively into the enterprise artificial intelligence (AI) model using natural language. AI agents capture
the knowledge according to a customisable enterprise ontology. SMEs teach the AI agents strategies and steps to encode the enterprise knowledge based on their experiences and the nature of their domains. Third-party algorithms, custom-trained large language models (LLMs), machine learning (ML) models, and other sources of knowledge and data can also be integrated with the enterprise AI model. As a result of this approach, comprehensive knowledge of the enterprise is captured in the enterprise AI model. hpad’s AI solutions are called hpad AI companions. They utilise the information available in the enterprise AI model to reason and solve problems holistically. There can be any number of them implemented in an enterprise, having different functionalities based on the use cases they support. Humans can have multimodal interactions with these AI companions depending on their requirements. hpad’s AI solutions also support creating simulations and digital twins of any aspect of the enterprise operations. AI companion for HAZOP hpad’s AI solutions can be used to reason and solve any enterprise problem holistically. Carrying out a HAZOP study is one such problem. A thorough HAZOP study is a time-consuming and resource-intensive process that requires a multidisciplinary team. Human factors, such as bias, fatigue, subjectivity, and lack of experience and skill, also pose challenges to the accuracy and consistency of the HAZOP study. The HAZOP also needs to be reviewed periodically as a policy or in case of any process changes. To overcome these challenges, the AI companion for HAZOP studies reasons based
Refining India
53
Powered by FlippingBook