Judge MAGNETO A) The report 1) organization, writing FULL POINTS 2) description of code and code listing * a code listing was provided * however, the code itself was only briefly described in "Model Solution". It may seem unreasonable, since there's little time, and the code is already described in the "Mathematical Modeling" document and you do refer to that. However, just so the report can stand alone, I would suggest some more words: "The code we have written is given estimated values for the many problem parameters, and initial values for the mosquito and human populations; for our runs, we always started with all healthy mosquitos, and with only one sick human in the village. Then we used a time-stepping method to solve the system of coupled equations that made up the model, very similar to the Forward Euler method, with a time step of one day (?), to update all the populations in accordance with the model. We saved the population values at each day so that we could make plots at the end showing the variations over time." ...and then continue with your text about "This method is an explicit method..." I would also explain to the judges why you made two versions of the basic code, one of which has a function form. 3) demonstrate understanding of the system and the model assumptions This information appears under "Model Design". Again, to make the report somewhat more independent of the "Mathematical Modeling" document, you might say something along the lines of: "A healthy human may be infected with malaria only through the bite of a mosquito that has previously fed on an infected human. Once infected, a human is ill for a certain time and then dies, becomes immune or simply recovers without immunity. Thus, the rate of new infections depends on the individual chance of being bitten by an infected mosquito, and the size of the healthy (but not immune) population. Similarly, the rate at which mosquitoes become infected with malaria depends on the chance of a bite occurring and the size of the infected human population. A further complication is that both mosquito and human populations have natural death and birth rates. A computer simulation can attempt to keep track of the changing values of the two populations over time, by implementing the model." 4) description of runs The instructions specifically said "Create the basic model and run it for 200 days. Examine the resulting breakdown of the human populations and the infected and uninfected mosquito population. Then, run the model for 1500 days and compare what happens to the trends in the disease." There is a discussion under "Sensitivity Analysis" which essentially includes this information, but as part of the sensitivity study. There should be an initial section called "Basic Model Runs" which simply computes the model with the given parameters for 200 and 1500 days and says something like "Look, everyone has been infected" or "The infected rate hit a plateau and then dropped to a very low value." 5) description of sensitivity runs This is fine. B) The codes The individual codes should each have a title line, such as "# extra_part3.py" so that a person looking at that file knows what it is called. After the title line, there should be a short explanation for every piece of code, explaining its purpose: # This code is used to examine the value of using treated or untreated nets. After the explanation, there might be a description of any new variables being used. This is done sometimes, but not for all the new variables: # treatedNets: an array which counts the number of treated nets in # use at each time step. The words "TODO:", which were supplied by the contest people, should NOT show up in the final code, because you DID it (we hope). Otherwise, a reader might assume this was an indication of some unfinished plan. 1) Correctness of base codes OK 2) Correctness of sensitivity runs OK C) The extra projects RELAPSE: 1) discussion in report Good explanation of the code changes, and good comment about difference in outcome. 2) correctness OK SEASONAL MOSQUITOES: Notice that, because we don't have plots for the basic runs of 200 and 1500 days, it's harder to compare the results here to the results with a constant mosquito rate. Also, I am puzzled by the 1500 day plot for the mosquitoes. I really wish we had the basic run plot for mosquitoes for comparison. I don't understand why the healthy population disappears as the infected population grows, especially since there are almost no infected villagers left. 1) discussion in report The code changes are clearly discussed. The human plots are surprising at first, but that's just because the time scale is dramatically changed. The mosquito plots are much more interesting. 2) correctness OK TREATED NETS: 1) discussion in report Figure (5) is not labeled Figure 5. Captions 5(a) and 5(b) run together in a confusing way. Caption 6 also seems confusing. Graph 5(a) uses a different scale from 5(b) and 6. If they all used the same scale, comparison would be easier. I see three plots, but I am not sure what is changing. Perhaps one change would be to label the curves using the word "strategy". Strategy 1: no one gets a net Strategy 2: everyone gets an untreated net Strategy 3: everyone gets a treated net Strategy 4: we start with half treated and half untreated, then new villagers get untreated nets. Now I would understand every labeled plot. The question asked for the cost of nets, but also for the effect that the nets would have on the malaria infection. I see that in the code there are computations that have been modified to account for the nets, but I don't see a discussion of this, nor do I see a plot over time of the Healthy/Sick/Immune Villagers and Healthy/Infected Mosquito rates for the various strategies. Comparing the cost of the strategies versus the corresponding infection rates would tell us what we get for our money. 2) correctness OK