There are a number of ways in which process design can go wrong. The best way to ensure it goes wrong seems in my experience to decide to dispense with the process engineer completely - they are after all expensive(but not, it turns out, as expensive as deciding to do without therm)
The second best is to assume that anyone with a degree in engineering is an engineer. Green engineering grads are not engineers yet. They require close supervision by a proper engineer. They may be cheaper than proper engineers, but this often proves a false economy.
The third best is to be unsure of who is responsible for process design. This either shades into the first example, or comes down to bad paperwork.
The fourth is to fail to manage the design process. Active management and formal quality control is required to produce a good design.
The fifth is to fail to manage the construction and commissioning process. Active management and formal quality control is also required to build a good plant .
The sixth is to fail to manage operation and maintenance properly. Operators are human, and left to their own devices, do what is easiest. I see a lot of dusty O+M manuals on site visits, unopened since the day the plant was started up.
To really screw things up usually requires a combination of these approaches. Figuring out which were responsible for plant failure is a fascinating task.