Research Topic 4: Semantic Data Interoperability Technologies across the Life Cycle of Process Plant Projects

Research Description

Poor data interoperability is a major issue in the process industry, hindering seamless data exchange and integration across project life cycles. Seamless data exchange is essential for effective communication and collaboration among stakeholders, with failure leading to delays, cost overruns, and data inconsistencies The core of the data interoperability problem lies in the diverse proprietary systems used by multiple stakeholders to process and exchange data that often lack standardized definitions and formats. To address this challenge, this research aims to develop semantic data interoperability technologies utilizing multi-modal language models. By enabling the seamless exchange and integration of diverse datasets from multiple, our goal is to improve data quality and increase the efficiency of work processes.

REU Research Plan:

REU students will be involved in developing data interoperability models for several key equipment and instruments included in process plants, under the guidance of postdoctoral researcher. This research involves three tasks: (1) identifying the terms to define the specifications of equipment and instruments from catalogs, manuals, and other resources; (2) matching disparate terms to define identical objects from various software; (3) leveraging multi-modal language models to transform data into a format that is compatible with various software templates.

Keywords: Data Interoperability, Data Exchange and Integration, Data Template, Multi-modal Language Model

Required Skills: Familiar with Large-Language Models

Undergraduate Degrees: Construction Management, Computer Science, Mechanical Engineering, Chemical Engineering

Faculty Advisor: Dr. David Jeong