Don’t Jump. Look Behind the Data.

Engineers from all over the world have the same interests and problems: how to increase production, how to save costs, how to improve operations, etc.  My recent presentation at the Middle East Process Engineering Conference (MEPEC) addressed these issues with a new approach to MEG reclamation.

While in Bahrain for the conference, I had the chance to attend Sameh Abdulqader Younis’ presentation on the “Non-Conventional Debottleneck Approach.” You know I love Sleuthing in Solid-Liquid Separation, and the Sherlock Holmesian approach, so I was intrigued immediately when the first point was: “Don’t Jump to Conclusions.”

Image source:  Billy Wilson Photography / Foter / CC BY-NC
Image source: Billy Wilson Photography / Foter / CC BY-NC

Sameh’s argument? For any process retrofitting and debottlenecking, we need to look at the process data and operating parameters — not as figures and numbers, but rather in a deeper and wider spectrum to see what is behind them. In certain cases, a piece of equipment may show a certain process constraint. Normally, a process engineer would conclude, “let us change it” and will start to look at the economics of making that move. Yet this approach, in Sameh’s words, should be considered “the Lazy Solution”.

An alternative action is for the professional process engineer to take a “Non-Conventional Approach” by looking at what is behind the process data and the operating parameters. You might be surprised to discover new ways of overcoming such an equipment bottleneck.

The presentation noted this approach has been successfully implemented in industries where a plant throughput was increased by around 3-5% with savings in operational costs.  In other cases, while certain operating data may not represent a bottleneck as the plant has reached its maximum throughput, looking at the data again might lead to a modification that would substantially increase the plant throughput. We have seen cases where throughput hit 25-35% over and above the design capacity, with almost zero capital investment!

How to proceed? Take these basic steps:

  1. Review the design of the unit/equipment and compare it to actual conditions.
  2. Check if support units and downstream processes are being correctly utilized.
  3. Check for similar units and compare design data for each, even if they are not at the same throughput.
  4. Check for internals design.
  5. Do not jump to conclusions that debottlenecking the downstream units will be the solution to overcome the main unit in question.

This entire approach is dependent on the human ability to solve a problem. The MEPEC presentation, and my own blogging, endorse breaking the paradigm and looking for the what’s behind the numbers/instruments/data to understand the issues. Follow Sherlock Holmes and “do not jump to conclusions”, ”learn to tell the crucial from the incidental” and “follow checklists.”  If I can be of assistance for your analysis, please reach out to me.  Good luck.