Fuzzy inference game approach to uncertainty in business decisions and market competitions

SpringerPlus

Table 1 Results of simulations of the untrained and trained agent in 2-player game

Agent moves			Untrained		Control expt		Trained
S/N	Green	Yellow	Winner	Payoff	Winner	Payoff	Winner	Payoff
1	3, 1, 1	2, 0, 3	Yellow	-289.3	Yellow	-192.9	Yellow	-305.0
2	0, 5, 0	1, 4, 0	Yellow	-99.8	Yellow	-66.6	Yellow	-142.2
3	0, 5, 0	0, 1, 4	Yellow	-704.8	Yellow	-469.9	Yellow	-747.2
4	4, 0, 1	4, 0, 1	Green	40.8	Green	61.3	Yellow	-8.2
5	1, 0, 4	2, 0, 3	Green	351.6	Green	527.5	Green	302.2
6	3, 1, 1	4, 0, 1	Yellow	-16.1	Yellow	-10.7	Yellow	-65.2
7	3, 0, 2	2, 1, 2	Green	136.8	Green	205.2	Green	94.9
8	3, 1, 1	3, 1, 1	Green	14.8	Green	22.34	Yellow	-34.2
9	1, 1, 3	1, 0, 4	Yellow	-22.0	Yellow	-14.7	Yellow	-63.9
10	2, 1, 2	1, 1, 3	Yellow	-52.7	Yellow	-35.2	Yellow	-94.8
11	3, 0, 2	2, 0, 3	Yellow	-26.7	Yellow	-17.9	Yellow	-68.8
12	0, 0, 5 (Geq)	0, 5, 0 (Leq)	Green	1054.5	Green	1581.8	Green	1012.0
13	0, 5, 0 (Leq)	0, 0, 5 (Geq)	Yellow	-863.8	Yellow	-575.9	Yellow	-906.2

From the table, the first column shows the serial numbers of the iterations, the second column contains player green’s strategies while the third column contains that of yellow. For example, in the fifth iteration, green’s strategy shows [1, 0, 4] this indicates how resources are allocated to strategy [C,W,M]: C = 1, W = 0 and M = 4. The forth column gives the winners for the untrained simulations while the fifth column gives the payoffs of those simulations. Column six and seven show the winners for the control experiment. The control experiments show the results where both players did not use fuzzy inference systems in playing the games. Column eight and nine show the winners for the trained simulations. These results show that the fuzzy player (Yellow) was able to win more than the competitor (Green) because he made use of the fuzzy inference system in making his business decisions. Also, it can be observed that the trained agent is able to perform better after training. The minus sign on yellow payoffs merely shows zero-sum. The strongest opponents (Geq) and weakest opponents (Leq) are shown in iterations 12 and 13 respectively.