2 ms·

I'm not sure this qualifies as an operations research problem. I mean, there is zero drawback from administering a vaccine. There is no need to pick between va

by rualca 6y ago

I'm not sure this qualifies as an operations research problem. I mean, there is zero drawback from administering a vaccine. There is no need to pick between vaccines. Vaccine cost is also not a constraint. Thus the optimal solution is, unequivocally, administer as many vaccines as possible, regardless of their effectiveness. The only thing that qualifies as a operations research problem is how to distribute a limited output of vaccines across a population in order to minimize the epidemic's spread and fatalities.

ggm 6y ago
The only thing that qualifies as a operations research problem is how to distribute a limited output of vaccines across a population in order to minimize the epidemic's spread and fatalities. Thats exactly what I am talking about. There are at least two choices: give to higher risk elderly people in care homes and care workers first, or give to frontline health staff first, and a pair of parallel choices of give one dose to as many as possible or give two full doses to the people who get the first dose, alongside which of the three currently qualified vaccines to give. This is precisely what I think is the O/R decision. What to do, with limited vaccine and multiple choices of how, and who to give it to. People are already arguing about what maximises benefit. Blair publicised a logical choice to give the maximum number of people one dose now, and give the second dose when more arrives. This invites the question if 80-90% coverage of fewer people is better or worse than 70% coverage of more people. He didn't think of it btw, he just publicised it. If this is called "the Blair choice" it will be a crying shame. There will be around 6 distinct vaccines soon. What if different vaccines give better response, than giving the same twice as a booster? (this is a possibility) People are trying to maximise benefit to the individuals, and to the economy. So its two-outcome maximising with at least 4 and very probably more like 8 or 10 distinct choices of outcome.

by rualca 6y ago

ggm 6y ago

The only thing that qualifies as a operations research problem is how to distribute a limited output of vaccines across a population in order to minimize the epidemic's spread and fatalities. Thats exactly what I am talking about. There are at least two choices: give to higher risk elderly people in care homes and care workers first, or give to frontline health staff first, and a pair of parallel choices of give one dose to as many as possible or give two full doses to the people who get the first dose, alongside which of the three currently qualified vaccines to give. This is precisely what I think is the O/R decision. What to do, with limited vaccine and multiple choices of how, and who to give it to. People are already arguing about what maximises benefit. Blair publicised a logical choice to give the maximum number of people one dose now, and give the second dose when more arrives. This invites the question if 80-90% coverage of fewer people is better or worse than 70% coverage of more people. He didn't think of it btw, he just publicised it. If this is called "the Blair choice" it will be a crying shame. There will be around 6 distinct vaccines soon. What if different vaccines give better response, than giving the same twice as a booster? (this is a possibility) People are trying to maximise benefit to the individuals, and to the economy. So its two-outcome maximising with at least 4 and very probably more like 8 or 10 distinct choices of outcome.