(155c) The Evolution of APC Project Methodology over the Past 25 Years

A Historical Perspective and Proposed Path Forward.

In the early 1990's, building an advanced control (APC) systems based on commercially available multivariable predictive control (MPC) software for a complex olefins or refining process cost well in excess of $1.5 million, and would require a team of 3 engineers to complete.

With the advent of the SmartStep automated step testing technology in 2000, the need for 24x7 engineering coverage was significantly reduced, and the project teams got down to 2 engineers. However, project teams discovered that the traditional APC project methodology formulated in the late 1980's and largely based on waterfall concepts was not well suited to fully exploit SmartStep (later versions were renamed Calibrate mode), and that significant methodology innovations were required to exploit the full benefit of step test automation and reduce project costs. The introduction of DMC3 in 2012 brought additional Model Identification capabilities that further reduced the man-hours needed to complete a DMC project: Constrained model identification, linearization, improved collinearity repair, and model quality analysis, which necessitated further methodology changes. Several large multi-national companies operating in the chemical, refining and olefins sectors experimented with methodology innovations, and found large cost savings by reducing the project teams from 3.5 to 1.5, reducing step test duration from 3 weeks to 1-1.5 weeks, and discovering that more accurate models lead to faster and easier controller commissioning and far less rework. Project costs came down between 300% and 900% between 2013-2023 period, compared with the early 1990's. However, adoption of the new methodology concepts was rather patchy, and many APC project teams largely remained with the traditional waterfall methodology and kept their project teams to 2 engineers, doing 3 week step tests and 3 week commissioning trips to site.

In this session we will consider these methodology innovations, and discuss why the combination of new technology and radical methodology changes lead to an order of magnitude cost reduction in some cases, what methodology innovations are most impactful, and discuss the path forward, specifically project execution using a hybrid approach combining remotely located project teams, with onsite resources who interact with the operators and onsite personnel. We will conclude with recommendation for future areas of innovation.