Talk:The Line of Least Resistance

From ESE205 Wiki
Jump to: navigation, search

Proposal Evaluation

List of objectives is unsatisfactory

Most of the objectives are not actual goals, but rather they are challenges to be solved in order to achieve a goal. For example, "Utilize simulation software to model the current system" is just a step to achieve a goal, not the goal of the project. Similar with "Become proficient in chosen software (MATLAB/SimEvents, Simul8, etc.)."

The only goal I can identify is "Model ways to improve current system based on information," which is vague both on the types of models and on the information that will be collected.

Collaboration with Arch officials

Although a collaboration would be interesting, it is most likely beyond the scope of a course such as ESE205. Collaborations usually take a long time to develop, and (as far as I understand) there is no indication that the Arch administration is actually interested in a new queue management system. Hence, the success of your project should not be associated to third parties whose interests are currently unknown.

Instead, a collaboration with the Arch administration would be feasible after you finish this semester's project. In that way, you would be able to demo a working system to them, which would in turn allow them to make a decision about your management system.

Data and models are undefined

The first step to model any system is to identify inputs, outputs, parameters, internal variables, and external variables. Based on those items one can begin identifying families of models that might help describing a system, and the kinds of estimation, decision, and control algorithms that can be used on those models.

None of that work is present in this proposal, which means that there is no concrete understanding of the problem, nor the kinds of solutions that can be offered. Identifying a suitable model will likely take between 2 and 4 weeks (maybe even longer). Thus, the timeline and objectives of this project must be revised accordingly.

Demo characteristics

The proposal mentions creating a program that will produce recommendations "for how to be most efficient at that time." First, the word "most" indicates that you are aiming for some form of optimal behavior, yet the definition of what is going to be optimized is not present in the proposal. Second, is this program going to be a Matlab script, which is probably not useful to an Arch official as they are unlikely to have a Matlab license, or is it going to be a smartphone app? If the answer is the latter, then this app must be accounted for in the Gantt chart.

This proposal needs a significant rewrite of the objectives, and a modification of the rest of the content accordingly. The revised version should be submitted no later than September 23 at 5pm.

-- Humberto 00:43, 20 September 2016 (CDT)