More on RLS and DMPR 2

More on RLS and DMPR

In the RLS process, after the leaders submit the list, there is nothing more that they can do. The result will come in a few minutes or even a few seconds if programmed commercial or custom made software is used. DMPR process could lead to the same outcome or different one. It will lead to the same outcome, if leaders follow the ranking order in picking. This is a safe approach.

But by changing the order, leaders could get a more favorable Coalition. This has also some risk. By doing that, they may end up with worse (according to their beliefs) composition than what following the ranking order in picking would result. Always, risk can lead to more reward but as the meaning of the word implies, it is not certain that it will not be worse.

It is a matter of examining the alternative outcomes, estimating the probabilities and finally taking a knowledgeable guess. There is no certainty in guess. In our examples, we present the lists of all the leaders. But the other leaders, do not know what the lists of the other "players" are. The will try to gues a) the lists of the other players b) how they may change the order of ranking while picking c) possible vetoes.

In RLS process, when leaders submit the lists, the other participants will have to submit their vetos as well. Before a Group is assigned to a Coalition, it should be checked if there is a veto for that. This can be done manually or automatically with software. In DMPR, the ranking lists will be secret. Vetoes will also be secret. They will be used, only if they are needed.

The order of picking or assigning is important in both DMPR and RLS. If the leaders can't agree on an order, this will have to be determined randomly. Suppose that the rounds for all of them were four or a multiple of that. The order could be in rotation as follows; C-I-A1-E2, I-A1-E2-C, A1-E2-C-I, E2-C-I-A1. Still, they would have to agree on the order of the first round or select it randomly.

Another way is to determine randomly, the order in each round. This makes the preparation a little harder and the time needed in DMPR between the rounds more. After each round (in DMPR), the players will need to reevaluate their strategy. In the examples, we assume that the order of picking is C-I-A1-E2 for all rounds since China and India will participate in only two rounds. But something different could be decided.

Leaders and the Groups that are close to them, will try to guess the other three lists and the possible vetoes. This is not easy at all. Most likely, there will be some differences between the actual lists and the lists that the other three players have guessed. It is like when in a poker game, the players try to guess what cards other players have. For the purposes of this example, we will keep it simple.

We will assume that the leaders have guessed correctly the lists of the other players. Also for simplisity, we will assume that there is no veto. First for A1 is A2. Suppose that A1 picks A2. C will get R4 in the second round. Although A2 is high in A1's list, it is not so high in other lists. In C's it is in eigth place and in E2's in seventh. In I's it is below tenth place. So there is no danger that other players will pick A2.

So A1 should not follow the ranking order in picking. Instead of picking first, the first Group on its list, A1 should pick the second, R4. That will be unavailable for C in the second round. The next available Group for C will be M2. Following the procedure that we have expalined, in the first two rounds, the Coalitions will be C-M3-M2, I-E1-R2, A1-R4-A2, E2-R1-M1.

In the next round, A1 will pick A3 and E2 will pick S2. Next available on A1's list is S4. Suppose that A1 picks S4. E2 will pick A3. S4 is in number fourteen of E2's list, the last choice. Again, A1 should change the ranking order and instead of picking S4 which is in eight place, it should pick R3 which is in ninth. In this way, R3 will be unavailable for E2. The next available Group for E2 is S1.

In the fifth and final round, A1 will pick S4 and E2 will pick S3. By changing the order of ranking in picking, A1 got; a) A2 and R4 b) S4 and R3. If the order was not changed, R4 would be in C's Coalition and R3 in E2's. Obviously, not only A1 can change the order. All leaders can do the same so that they can improve the composition of their Coalition. A1 was only used as an example. The five rounds are shown in the table below.

By using DMPR and RLS we have come with four arrangements (combinations) in the two previous articles.

• 1) C-M3-R2   I-S1-S3   A1-A2-A3-R3-R4-S4   E2-E1-M1-M2-R1-S2   not India
• 2) C-M3-R4   I-E1-R2   A1-A2-A3-M2-S4-S1   E2-M1-R1-R3-S2-S3   not veto
• 3) C-M3-R4   I-E1-R2   A1-A2-A3-R3-R4-S1   E2- R1-M1-S2-S3-S4   veto
• 4) C-M3-M2  I-E1-R2   A1-A2-A3-R3-R4-S4   E2-R1-M1-S1-S2-S3   change order (only DMPR)
  

  ---

In some cases, there is geographic proximity. In third and fourth arrangement, E2's Coalition has two Subcoalitions with geographic proximity, E2- R1-M1, S2-S3-S4 and E2-R1-M1, S1-S2-S3. There can be EXCHANGES of Groups between Coalitions, if there is MUTUAL AGREEMENT. Having geographic proximity is one reason but there can be other reasons for exchanges.

Suppose that in the first arrangement, A1 and E2 exchange S4 fro R1. In A1's Coalition, there will be two Subcoalitions with geographic proximity, A1-A2-A3 and R1-R3-R4. In E2's there will be geographic proximity in E2-E1-M1-M2 and in S2-S4. Whether these are next to each other, depends on S2's composition which we do not know yet. Even if they are not next to each other, they will not be very far. There are other exchanges that could be made for geograhpic (spatial) proximity.

Obviously, the actual lists of the leaders will be different than the ones presented which were just examples. We do not know which criteria leaders will use to compile their lists. Perhaps they will rank Groups that are close to them higher. In that case, there will be much more spatial proximity in the four Coalitions that will result with DMPR/RLS process. With a few exchanges, there can be even more spatial proximity.

If we add one veto in the DMPR process explained above, it will get even more complicated. As it is obvious to everyone, it is already too chalenging and complicated. If leaders have to decide about grouping, it will be even more challenging complicated. The grouping provided is a very good one. It works well for forming Coalitions and following all the processes described. Another grouping may not work that well.

If it is hard already to reach an agreement about the arrangement of Groups in Coalitions, it will be even harder if the decision about grouping is added. Nevertheless, decision makers could try other groupings, with the assistance of data analysts. Data analysts should know the basics of game theory. If more advanced knowledge of game theory is required, data analysts may know experts in game theory and bring them in the teams that assist the leaders.