w201rdada / portfolio-hmorgan143

portfolio-hmorgan143 created by GitHub Classroom
0 stars 0 forks source link

Comments from Grace #5

Open graceyqlin opened 6 years ago

graceyqlin commented 6 years ago

Hi Hannah,

Thank you for sharing your idea! I like the layout of your idea. It clearly states your mission, the market demand, your solutions, and possible obstacles to optimization. It' very interesting to learn how smart chemicals can help to create models to perfect formulation of chemicals!

As you mentioned, paper making is a very complicated process. I personally am not very familiar with this industry. When a mill produce paper, does it have specific formula it needs to use? Is there any testing machines and techniques available for quality control? Are you just changing the formula for the chemicals or you need to adjust the whole production line? Do mills have database which can provide you with the data like pulp consistency, temperature, fiber content, and moisture content?

Overall, your project can be a game changer for the not only paper making but also the entire chemical industry! Thank you again for sharing your idea!

Grace

sioharr commented 6 years ago

As you anticipated, I had no idea how complicated paper is! I like your first paragraph because it helps me understand why this is a big deal.

In the second section, I think the reader is supposed to have a good understanding of your approach. I am a little confused. Will the variables you mentioned in the first paragraph be the primary variables used in machine learning (in addition to the actual chemical formula)? Perhaps you could also consider how you determine you have the right chemical cocktail? At the end of the process, is paper graded? How will you manage quality vs cost? Are you prioritizing one over the other?

Who is the audience? Usually, when I have watched people at work build their case, they do not highlight their obstacles. Could this section actually be flipped to be a competitive advantage of your product? Just off the top of my head... "Other chemical optimization methods do not consider the unique needs of your mill. We will partner closely with you to create a customized solution that will include potential plant optimization upgrades that will provide benefits across your operation". I think this part also belongs in the middle, solution section.

Then, I think you should finish strong with the impacts of your project. You allude to impact in the beginning, but can this tool impact more than just costs? Will it also improve quality? Will it impact a company's position in the paper market? Or is this something paper mills NEED to do to be around in business in 5 years? Do these chemicals have an environmental impact? Will some of the infrastructure upgrades that your project requires have far reaching benefits?

I am jealous of your deep knowledge on this subject. It seems like a great fit for machine learning and a fairly untapped data science field! Thanks for making this complicated subject accessible to non-paper people!

hmorgan143 commented 6 years ago

thank you both for your comments! they are super helpful (especially in making sure i am explaining myself to non-industry folks appropriately).