When I moved to Austin TX in 1997, there was a semiconductor boom happening in Texas. I soon was introduced to projects at places like Samsung, Motorola, and Texas Instruments. Among the many differences this industry has from the paper industry is the speed with which facilities are built, upgraded, and decommissioned.
The nature of the semiconductor business is that Moore's law dictates that current technology has a short shelf life, so the current process equipment also has a short shelf life. Motorola was in the midst of downsizing when I arrived in Texas, and I became project manager for decommissioning projects at some of their fabs. In the semiconductor industry, new process nodes ramp rapidly, often reaching peak production within a year or two. The manufacturing equipment is then continually upgraded or replaced as newer generations arrive, while the fab itself is often renovated and reused for decades. Compared with traditional process industries, semiconductor manufacturers are far less likely to leave obsolete production equipment abandoned in place for years.
In industries like paper, our facilities are built to make money for 50 years or more. When mills abandon batch digesters for Kamyr, those batch digesters may be left hanging and dormant. Not only the digesters, but also instrumentation, wiring, piping, and other process equipment is abandoned to rust and depreciate. The same goes for any other unit operation.
It has often puzzled me to walk through abandoned unit operations for the following reasons:
- As soon as that process is shut down, you have working equipment that can be reused. There are lots of valves, motors, sensors, DCS/SCADA components, piping, cable, and other equipment and materials. The longer it sits in the snow and rain, cold and heat, dirt and dust, the less useful it will become. There are firms that will buy that equipment and warehouse it for resale, so you don't have to maintain your own inventory.
- Leaving all that space occupied by a "fallow field" seems nutty. If the abandoned process is replaced by a new one, the new process gets crammed into an awkward position requiring long and byzantine piping runs.
- The bigger danger is the safety risk. It becomes increasingly difficult to know what is still live and what has been abandoned when piping and cable from active systems run through obsolete process areas. The building itself can be a hazard if it is no longer structurally sound. I have seen buildings with brick walls that are crumbling and trees growing through the roof. You don't want to be walking next to such a building when it decides to collapse.
There are several legitimate counterarguments:
- Full decommissioning and demolition is expensive. Having done this, it is not a trivial pursuit. It requires careful planning to make sure you don't accidentally shut down a live process or injure someone.
- There is also the matter of disposing of materials. While there are firms that may buy used equipment for resale, or buy scrap metal for recycling and resale, concrete and brick need to be disposed of on your own.
- It is cheaper to keep everything in place and use it as your boneyard. That way, when you need a pressure sensor you can check your inventory and find the replacement in your facility without having to buy it back.
- Decommissioning does not require demolition. It is possible to mark all the piping and cables so that is it clear what is live and dead.
- Vendors like to sell new stuff, so having a larger market for reused materials may depress the market for those vendors.
- I remember asking a mill manager about why there was so much abandoned equipment in his mill. I confess I got lost in his answer, but it seemed to be based on perverse incentives of the tax code and the way depreciation works.
On balance, I think it is very wasteful and unsustainable for industry to abandon in place. What do you think? I realize there are sound economic reasons why many companies choose to abandon equipment in place rather than remove it. Even so, I can't help wondering whether we've accepted a practice that is neither economically optimal nor environmentally sustainable. Have I overlooked something, or is it time for industry to rethink how we manage obsolete process equipment? I would appreciate your comments. Feel free to reply to PatJDixon@DPAS-INC.com or on social media when this article is posted.
