Chemical plant design Hypothetical scenario 2

[Sanchit1331]

Let's say you are given control over a fairly simple chemical plant manufacturing a resin product. There are no documents or any material which can help you in understanding the process. You deduct the following by just observing the production flow.

1. Workers are loading feedstock material (waste sludge) in a makeshift melting vessel which requires 12-14 hours to "melt" the material. No standard recipe exists and workers load the feedstock material purely on trial-error basis.
2. The melted material is poured on steel molds resembling ingots where they cool at room temperature and solidify into a wax type substance. This takes about 2-3 hours.
3. Those ingots are considered final product and packed into corrugated boxes for shipping.
4. In another parallel process, ingots are crushed into smaller pieces and feed into an extruder system for manual pelltizing
There are two extruders.

One production cycle takes around 24 hours. An average of 12 workers are required for optimum efficiency.
What would you, as the owner and chemical expert, would do to maximize production and reduce job costing.

 

[MJCronin]

I smell homework ....

MJCronin
Sr. Process Engineer

 

[Sanchit1331]

It's not homework. It's a very gross oversimplification of a problem I am currently going through.

 

[Compositepro]

A common method used in chemical plants and refineries is "debottlenecking". You look at each step in the process and determine which step is limiting production output. You improve this one step until it is no longer the bottleneck. Then you move on to improving the new bottle-neck.

In your case it seems that the bottleneck is the long batch melting time and perhaps the number of ingot molds required to handle a large batch. Can you turn this into a semi-continuous process by adding material to the melting pot while drawing off melted material? Cold material should sink and molten material should rise to the top. Then you can be continuously and simultaneously melting and producing ingots.

 

[Sanchit1331]

Thanks compositepro for your post. Your advice makes sense, a major bottleneck is the long time of the reaction and the limited number of ingot molds. Also the puring of molten material into the molds is done using "buckets" which itself takes around 1-2 hours to fill all the molds.

Will you be kind enough to point me any case study or any reference material for the operation you just mentioned so that I can analyze it further ?

 

[Latexman]

Create a standard recipe. This may eliminate or greatly reduce the trial-error.

Good luck,
Latexman

Engineers helping Engineers

 

[mk3223]

1.

No standard recipe exists and workers load the feedstock material purely on trial-error basis.

It looks like a secret receipt which has no fixed composition, How can you control the quality of the final production?

2. May consider to replace some labor work with automated process with AI & robot, which may improve the operation safety and efficient around the clock. Suggest to have an Industrial or Manufacturing Expert to help you for the improvement proposal.

 

[Sanchit1331]

It looks like a secret receipt which has no fixed composition, How can you control the quality of the final production?

The feedstock material availablility is random and therefore the process is subjected to trial and error whenever new supply comes in. Also due to occasional incomplete meltdown, left over residue in the reactor and random hardness of the feedstock material, it's hard to standardise a recipe. Any pointers on this ?