Talk:Grocery Guru

From ESE205 Wiki
Revision as of 13:45, 7 February 2017 by Hgonzale (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Proposal Review

Strengths

  • Idea is clearly explained in the proposal.
  • Main challenges are identified in objectives and challenges.
  • There is a clear problem that this app will address, hence we know what to expect as results.

Weaknesses

The fundamental problem this proposal addresses is related to the famous Traveling Salesman Problem, a classic problem studied in computer science theory and operations research. Depending of the number of items you want to purchase, and the available total number of items in the supermarket, finding optimal solutions for this problem can be extremely hard (both in terms of time and computation).

On the other hand, this proposal hinges on the availability of a real-world database listing all the items in a supermarket, their location, possibly price, and other properties such as whether is frozen. Obtaining that data will be hard, unless serious simplifications are applied (e.g., grouping all the cereals as one, avoid listing all the types of cheese or yogurt, etc.) which potentially render the app useless. Building the database by yourselves is very time intensive, thus you are unlikely to do it in one or two weeks as suggested by your Gantt chart. Moreover, there are better ways to use your time than roaming through Schnucks creating a huge list of all the items you can find.

Unless you can obtain, in the next week or two, the database you need for this project, I fail to see how this idea can succeed. If you propose to stick with the plan of building the database yourselves, then the proposal should have a clear plan on how to do it, including the type of database you will build and where you are going to host it (a large Excel spreadsheet is insufficient for this amount of data, and you won't be able to extract the data to run your optimization algorithm from such a spreadsheet either).

Other comments:

  • Creating easy-to-use user interfaces is challenging. Your proposal should at least mention what are your fundamental ideas of the features in the app.
  • Where are you going to host the database, and what kind of computational platform do you need to solve your traveling salesman problem? A smartphone is incapable of doing either for a reasonably large supermarket, hence your budget must consider cloud hosting of some sort.
  • How are you going to demo your project? A simulation is not enough, the demo must be interactive.


Humberto 13:35, 7 February 2017 (CST)