US2003208514A1PendingUtilityA1

Methods and apparatus for decision making

Priority: Apr 30, 2002Filed: Apr 30, 2003Published: Nov 6, 2003
Est. expiryApr 30, 2022(expired)· nominal 20-yr term from priority
G06N 7/01G06F 3/0481
42
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

There is disclosed a multiple criteria decision analysing method in which a plurality L of basic criteria are assessed in order to a general criterion, comprising the steps of: making an assessment {(K m,l ,γ m,l ),m=1, . . . , M} of the l th basic criteria under a set of grades {K m,l ,m=1, . . . , M}; and transforming the assessment {(K m,l ,γ m,l ),m=1, . . . ,M} to an assessment {(H n ,β n,l ),n=1, . . . ,N} of the general criterion under a set of grades {H n ,n=1, . . . N} using the matrix equation: [ β 1 , l β 2 , l ⋮ β N , l ] = [ α 1 , 1 α 1 , 2 … α 1 , M α 2 , 1 α 2 , 2 … α 2 , M ⋮ ⋮ ⋰ ⋮ α N , 1 α N , 2 … α N , M ] = [ γ 1 , l γ 2 , l ⋮ γ M , l ] wherein: H n is the n th grade for assessment of the general criterion; K m,l is the m th grade for assessment of the l th basic criterion; α n,m is the m th degree to which K m,l implies H n ; γ m,l is the degree to which the l th basic criterion is assessed to K m,l ; β n,l is the degree to which the th basic criterion is assessed to H n ; and ∑ n = 1 N  α n , m = 1

Claims

exact text as granted — not AI-modified
What is claimed is:  
     
         1 . A multiple criteria decision analysis method in which a plurality L of basic criteria are assessed in order to a general criterion, comprising the steps of: 
 making an assessment {(K m,l ,γ m,l ),m=1, . . . ,M} of the l th  basic criteria under a set of grades {K m,l , Im=1, . . . ,M}; and    transforming the assessment {(K m,l γ m,l ),m=1, . . . , M} to an assessment {(H n ,β n,l ),n=1, . . . ,N} of the general criterion undera set of grades {H n , n=1, . . . ,N} using the matrix equation:              [           β     1   ,   l                 β     2   ,   l               ⋮             β     N   ,   l             ]     =       [           α     1   ,   1             α     1   ,   2           ⋯         α     1   ,   M                 α     2   ,   1             α     2   ,   2           ⋯         α     2   ,   M               ⋮       ⋮       ⋰       ⋮             α     N   ,   1             α     N   ,   2           ⋯         α     N   ,   M             ]     =     [           γ     1   ,   l                 γ     2   ,   l               ⋮             γ     M   ,   l             ]                         wherein: 
 H n  is the n th  grade for assessment of the general criterion;  
 K m,l  is the m th  grade for assessment of the l th  basic criterion;  
 α n,m  is the degree to which K m,l  implies H n ;  
 γ m,l  is the degree to which the l th  basic criterion is assessed to K m,l ;  
 β n,l  is the degree to which the th basic criterion is assessed to H n ; and  
             ∑     n   =   1     N          α     n   ,   m         =   1.                   
   
     
     
         2 . A method according to  claim 1  in which the values of α n,m (m=1, . . . , M and n=1, . . . ,N) are assigned by a decision maker.  
     
     
         3 . A method according to  claim 2  in which a grade K m,l  implies a grade H n  to a degree of α n,m  wherein m=1, . . . ,M.  
     
     
         4 . A method according to  claim 1  in which the values of α n,m (m=1, . . . ,M and n=1, . . . ,N) are determined using the following equations:  
       
         
           
             
               
                 
                   α 
                   
                     n 
                     , 
                     m 
                   
                 
                 = 
                 
                   
                     
                       u 
                        
                       
                         ( 
                         
                           H 
                           
                             n 
                             + 
                             1 
                           
                         
                         ) 
                       
                     
                     - 
                     
                       u 
                        
                       
                         ( 
                         
                           K 
                           
                             m 
                             , 
                             l 
                           
                         
                         ) 
                       
                     
                   
                   
                     
                       u 
                        
                       
                         ( 
                         
                           H 
                           
                             n 
                             + 
                             1 
                           
                         
                         ) 
                       
                     
                     - 
                     
                       u 
                        
                       
                         ( 
                         
                           H 
                           n 
                         
                         ) 
                       
                     
                   
                 
               
               , 
               
                 
                   α 
                   
                     
                       n 
                       + 
                       1 
                     
                     , 
                     m 
                   
                 
                 = 
                 
                   1 
                   - 
                   
                     α 
                     
                       n 
                       , 
                       m 
                     
                   
                 
               
               , 
             
           
           
           
               
           
         
         α i,m =0(i=1, . . . ,N,i≠n,n+1)  
         if u(H n )≦u(K m,l )≦u(H n+1 ) for n=1, . . . ,N−1;m=1, . . . ,M where u(H n ) and u(K m,l ) are utilities of H n  and K m,l , respectively.  
       
     
     
         5 . A method according to  claim 4  in which at least one of u(H n ) and u(K m,l ) are estimated by a decision maker.  
     
     
         6 . A method according to  claim 4  in which at least one of u(H n ) and u(K m,l ) are determined using one or more of the equations:  
       
         
           
             
               
                 
                   u 
                    
                   
                     ( 
                     
                       H 
                       n 
                     
                     ) 
                   
                 
                 = 
                 
                   
                     
                       
                         n 
                         - 
                         1 
                       
                       
                         N 
                         - 
                         1 
                       
                     
                      
                     
                         
                     
                      
                     for 
                      
                     
                         
                     
                      
                     n 
                   
                   = 
                   1 
                 
               
               , 
               … 
                
               
                   
               
               , 
               
                 N 
                  
                 
                     
                 
                  
                 if 
                  
                 
                     
                 
                  
                 
                   H 
                   
                     n 
                     + 
                     1 
                   
                 
               
             
           
           
           
               
           
         
       
       is preferred to H n ; and  
       
         
           
             
               
                 
                   u 
                    
                   
                     ( 
                     
                       K 
                       
                         m 
                         , 
                         l 
                       
                     
                     ) 
                   
                 
                 = 
                 
                   
                     
                       
                         m 
                         - 
                         1 
                       
                       
                         M 
                         - 
                         1 
                       
                     
                      
                     
                         
                     
                      
                     for 
                      
                     
                         
                     
                      
                     m 
                   
                   = 
                   1 
                 
               
               , 
               … 
                
               
                   
               
               , 
               
                 M 
                  
                 
                     
                 
                  
                 if 
                  
                 
                     
                 
                  
                 
                   K 
                   
                     
                       m 
                       + 
                       1 
                     
                     , 
                     l 
                   
                 
               
             
           
           
           
               
           
         
       
       is preferred to K m,l .  
     
     
         7 . A method according to  claim 1  in which: 
 a basic criterion is assessed using the set {(k j , p j ),=1, . . . ,P}, where k j  is a number and p j  is the probability of the basic criterion taking the number k j ; and  
 the values of γ m,l  in the assessment {(K m,l , γ m,l ),m=1, . . . ,M} are calculated using the equation  
             γ     m   ,   l       =         ∑     j   =   1     P            S     m   ,   j            p   j                   for                 m       =   1       ,   …              ,   M               wherein   :     S     m   ,   j         =         K       m   +   1     ,   l       -     k   j           K       m   +   1     ,   l       -     K     m   ,   l             ,       S       m   +   1     ,   j       =     1   -     S     m   ,   j           ,         S     i   ,   j       =   0     ;                     
 m=M, . . . ,M−1; j=1, . . . ,P; i=1, . . . ,M,i≠m,m+1 if K m,i ≦k≦K m+1,l    
 
     
     
         8 . A multiple criteria decision analysis method in which a plurality L of basic criteria are assessed in order to assess a general criterion, comprising the steps of: 
 assigning weights W i (i=1, . . . L) to the L basic criteria;    normalising the weights using the equations                ω   i     =         W   i         ∑     j   =   1     L                     W   j                         (       i   =   1     ,                …                 L       )         ;                     determining β n,l , wherein β n,i  is the degree to which the i th  basic criterion is assessed to H n , and H n  is the n th  grade for assessment of the general criterion, the general criterion being assessed into N grades;    calculating weighted degrees of belief n i from the equation      m   n,i   =ω   i β n,i )=ω i β n,i ,( n= 1 , . . . , N; i=. . . ,  1);and    calculating a remaining probability mass m H,i  from the equation              m     H   ,   i       =     1   -       ∑     n   =   1     N                       m     n   ,   i                         (       i   =   1     ,              …              ,   L     )     .                             
     
     
         9 . A method according to  claim 8  in which m H,i  is decomposed into {overscore (m)} H,i  and {tilde over (m)} H,i , wherein:  
         m   H,i   ={overscore (m)}   H,i   +{tilde over (m)}   H,i ;  {overscore (m)}   H,i =1−ω i ;and              m   ~       H   ,   i       =           ω   i          (     1   -       ∑     n   =   1     N                     β     n   ,   i           )                     for                 i     =   1       ,              …              ,     L   .                     
     
     
         10 . A method according to  claim 9  in which m n,i ,{overscore (m)} H,i  and {tilde over (m)} H,i (i=1, . . . ,L) are aggregated into combined probability masses I n,L ,{overscore (I)} H,L  and Ĩ H,L , respectively, using equations i) to ix) in a recursive manner  
       I n,1=m   n,l (n=1,2, . . . ,N)  i) I H,l =m H,l   ii) Ĩ H,l ={tilde over (m)} H,l   iii) {overscore (I)} H,l ={overscore (m)} H,l   iv)                K     i   +   1       =       [     1   -       ∑     t   =   1     N            ∑       j   =   1       j   ≠   t       N            I     t   ,   i            m     j   ,     i   +   1                 ]       -   1               v   )                       I n,i+1   =K   i+1   [I   n,i   m   n,i+1   +I   H,i   m   n,i+1   +I   n,i   m   H,i+1 ]( n= 1,2 . . . , N)  vi) Ĩ H,i+1   =+K   i+1   [Ĩ   H,i   {tilde over (m)}   H,i+1   +{overscore (I)}   H,i+1   {tilde over (m)}   H,i+1   +Ĩ   H,i+1   {overscore (m)}   H,i+1 ]  vii) {overscore (I)} H,i+1   =K   i+1   [{overscore (H)}   H,i+1   {overscore (m)}   H,i+1 ]  viii)  I   H,i+1   ={overscore (I)}   H,i+1   +{overscore (I)}   H,i+1   ix)  i={1,2, . . . ,L−1}   
     
     
         11 . A method according to  claim 10  in which combined degrees of belief β n  and β H  are generated using the equations:  
       
         
           
             
               
                 
                   
                     
                       
                         β 
                         n 
                       
                       = 
                       
                         
                           
                             
                               I 
                               
                                 n 
                                 , 
                                 L 
                               
                             
                             
                               1 
                               - 
                               
                                 
                                   I 
                                   _ 
                                 
                                 
                                   H 
                                   , 
                                   L 
                                 
                               
                             
                           
                            
                           
                               
                           
                            
                           n 
                         
                         = 
                         
                           1 
                           , 
                           2 
                           , 
                           
                               
                           
                            
                           … 
                         
                       
                     
                      
                     
                         
                     
                     , 
                     N 
                   
                 
               
               
                 
                   
                     
                       β 
                       H 
                     
                     = 
                     
                       
                         
                           I 
                           ~ 
                         
                         
                           H 
                           , 
                           L 
                         
                       
                       
                         1 
                         - 
                         
                           
                             I 
                             _ 
                           
                           
                             H 
                             , 
                             L 
                           
                         
                       
                     
                   
                 
               
             
           
           
           
               
           
         
       
       wherein β n  is a degree of belief to which the general criterion is assessed to the n th  grade H n  and β H  is a remaining degree of belief which is not assigned to any specific grade.  
     
     
         12 . A method according to  claim 11  wherein each grade H n+1  is more favourable than H n  and performance indicators of a general criterion are generated using the equations:  
       
         
           
             
               
                 
                   
                     
                       u 
                       max 
                     
                     = 
                     
                       
                         
                           ∑ 
                           
                             n 
                             = 
                             1 
                           
                           
                             N 
                             - 
                             1 
                           
                         
                          
                         
                             
                         
                          
                         
                           
                             β 
                             n 
                           
                            
                           
                             u 
                              
                             
                               ( 
                               
                                 H 
                                 n 
                               
                               ) 
                             
                           
                         
                       
                       + 
                       
                         
                           ( 
                           
                             
                               β 
                               N 
                             
                             + 
                             
                               β 
                               H 
                             
                           
                           ) 
                         
                          
                         
                           u 
                            
                           
                             ( 
                             
                               H 
                               N 
                             
                             ) 
                           
                         
                       
                     
                   
                 
               
               
                 
                   
                     
                       u 
                       min 
                     
                     = 
                     
                       
                         
                           ( 
                           
                             
                               β 
                               1 
                             
                             + 
                             
                               β 
                               H 
                             
                           
                           ) 
                         
                          
                         
                           u 
                            
                           
                             ( 
                             
                               H 
                               1 
                             
                             ) 
                           
                         
                       
                       + 
                       
                         
                           ∑ 
                           
                             n 
                             = 
                             2 
                           
                           N 
                         
                          
                         
                             
                         
                          
                         
                           
                             β 
                             n 
                           
                            
                           
                             u 
                              
                             
                               ( 
                               
                                 H 
                                 n 
                               
                               ) 
                             
                           
                         
                       
                     
                   
                 
               
               
                 
                   
                     
                       u 
                       avg 
                     
                     = 
                     
                       
                         
                           u 
                           max 
                         
                         + 
                         
                           u 
                           min 
                         
                       
                       2 
                     
                   
                 
               
             
           
           
           
               
           
         
       
       and wherein u max , u min  and u avg  are the best possible, worst possible and average performance indicators respectively, and u(H n )(n=1, . . . N) is optionally defined by  
       
         
           
             
               
                 u 
                  
                 
                   ( 
                   
                     H 
                     n 
                   
                   ) 
                 
               
               = 
               
                 
                   n 
                   - 
                   1 
                 
                 
                   N 
                   - 
                   1 
                 
               
             
           
           
           
               
           
         
       
     
     
         13 . A method according to  claim 8  in which the values of β n,i  are determined using a method according to  claim 1 .  
     
     
         14 . A carrier medium storing a computer program, which computer program performs a method according to  claim 1 .  
     
     
         15 . A computer adapted to perform a method according to  claim 1 .  
     
     
         16 . A carrier medium storing a computer program, which computer program performs a method according to  claim 8 .  
     
     
         17 . A computer adapted to perform a method according to  claim 8.

Join the waitlist — get patent alerts

Track US2003208514A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.