3.1 基于模糊熵的区间直觉模糊VIKOR方法
3.1.1 问题描述
设某多属性决策问题有 m m m个方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m),组成方案集 Y = ( Y 1 , Y 2 , ⋯ , Y m ) Y=\left(Y_{1},Y_{2},\cdots,Y_{m}\right) Y=(Y1,Y2,⋯,Ym),评价每个方案的属性(或指标)为 G j ( j = 1 , 2 , ⋯ , n ) G_{j}\left(j=1,2,\cdots,n\right) Gj(j=1,2,⋯,n),记属性集为 G = { G 1 , G 2 , ⋯ , G n } G=\left\{ G_{1},G_{2},\cdots,G_{n} \right\} G={G1,G2,⋯,Gn}。如果 F ~ i j = ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ ( i = 1 , 2 , ⋯ , m ; j = 1 , 2 , ⋯ , n ) \tilde{F}_{ij} = \left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle \left(i=1,2,\cdots,m;j=1,2,\cdots,n\right) F~ij=⟨[μijL,μijU],[νijL,νijU]⟩(i=1,2,⋯,m;j=1,2,⋯,n)为区间直觉模糊集,表示方案满足属性 G j ∈ G G_{j} \in G Gj∈G和不满足属性 G j ∈ G G_{j} \in G Gj∈G的程度,且 0 ≤ μ i j + ν i j ≤ 1 0 \leq \mu_{ij}+\nu_{ij} \leq 1 0≤μij+νij≤1,矩阵 F = ( ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ ) m × n F = \left(\left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle\right)_{m×n} F=(⟨[μijL,μijU],[νijL,νijU]⟩)m×n为该多属性决策问题的区间直觉模糊决策矩阵。现在的问题是依据区间直觉模糊决策矩阵 F F F,如何得到一个有效的决策分析方法来对所有方案进行优劣排序。
3.1. 基于区间模糊熵的属性权重确定方法
设 A ~ = { ⟨ x i , [ μ A L ( x i ) , μ A U ( x i ) ] , [ ν A L ( x i ) , ν A U ( x i ) ] ⟩ ∣ x i ∈ X , i = 1 , 2 , ⋯ , n } \tilde{A} = \left\{ \left\langle x_{i},\left[\mu_{AL}\left(x_{i}\right),\mu_{AU}\left(x_{i}\right)\right], \left[\nu_{AL}\left(x_{i}\right),\nu_{AU}\left(x_{i}\right)\right] \right\rangle | x_{i} \in X, i=1,2,\cdots,n\right\} A~={⟨xi,[μAL(xi),μAU(xi)],[νAL(xi),νAU(xi)]⟩∣xi∈X,i=1,2,⋯,n}为论域 X X X上的区间直觉模糊数,则
E ( A ~ ) = 1 n ∑ i = 1 n cos π ( ∣ μ A L 2 ( x i ) − ν A L 2 ( x i ) ∣ + ∣ μ A U 2 ( x i ) − ν A U 2 ( x i ) ∣ ) 2 (3.1) \color{red} { E\left(\tilde{A}\right) = \frac{1}{n}\sum_{i=1}^{n}\cos\frac{\pi\left( \left|\mu_{AL}^2\left(x_{i}\right) - \nu_{AL}^2\left(x_{i}\right)\right| + \left|\mu_{AU}^2\left(x_{i}\right) - \nu_{AU}^2\left(x_{i}\right)\right| \right)}{2} \tag{3.1} } E(A~)=n1i=1∑ncos2π(∣∣μAL2(xi)−νAL2(xi)∣∣+∣∣μAU2(xi)−νAU2(xi)∣∣)(3.1)
是一个区间直觉模糊熵。 也可以改写为
E ( A ~ ) = 1 n ∑ i = 1 n cos π ( ∣ μ A L ( x i ) − ν A L ( x i ) ) ( 1 − π A L ( x i ) ∣ + ∣ μ A U ( x i ) − ν A U ( x i ) ) ( 1 − π A U ( x i ) ∣ ) 2 (3.2) \color{red} { E\left(\tilde{A}\right) = \frac{1}{n}\sum_{i=1}^{n}\cos\frac{\pi\left( \left|\mu_{AL}\left(x_{i}\right) - \nu_{AL}\left(x_{i}\right)\right)\left(1-\pi_{AL}\left(x_{i}\right)\right| + \left|\mu_{AU}\left(x_{i}\right) - \nu_{AU}\left(x_{i}\right)\right)\left(1-\pi_{AU}\left(x_{i}\right)\right| \right)}{2} \tag{3.2} } E(A~)=n1i=1∑ncos2π(∣μAL(xi)−νAL(xi))(1−πAL(xi)∣+∣μAU(xi)−νAU(xi))(1−πAU(xi)∣)(3.2)
从式 ( 3.2 ) (3.2) (3.2)不难看出,区间直觉模糊熵 E ( A ~ ) E(\tilde{A}) E(A~)不仅考虑了隶属度与非隶属度的偏差 ( μ A L ( x i ) − ν A L ( x i ) (\mu_{AL}\left(x_{i}\right) - \nu_{AL}\left(x_{i}\right) (μAL(xi)−νAL(xi), ( μ A U ( x i ) − ν A U ( x i ) (\mu_{AU}\left(x_{i}\right) - \nu_{AU}\left(x_{i}\right) (μAU(xi)−νAU(xi),而且考虑了犹豫度 π A L ( x i ) \pi_{AL}\left(x_{i}\right) πAL(xi)、 π A U ( x i ) \pi_{AU}\left(x_{i}\right) πAU(xi)的信息。
在多属性决策问题中,由于各备选方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)在属性 G j G_j Gj的评价值 F ~ i j = ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ \tilde{F}_{ij} = \left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle F~ij=⟨[μijL,μijU],[νijL,νijU]⟩为区间直觉模糊数,根据式 ( 3.2 ) (3.2) (3.2)可得属性 G j ( j = 1 , 2 , ⋯ , n ) G_j\left(j=1,2,\cdots,n\right) Gj(j=1,2,⋯,n)的区间直觉模糊熵 E j E_j Ej为
E j = 1 m ∑ i = 1 m cos π ( ∣ ( μ i j L − ν i j L ) ( 1 − π i j L ) ∣ + ∣ ( μ i j U − ν i j U ) ( 1 − π i j U ) ∣ ) 2 , j = 1 , 2 , ⋯ , n (3.3) \color{red} { E_{j} = \frac{1}{m}\sum_{i=1}^{m}\cos\frac{\pi\left( \left|\left(\mu_{ijL} - \nu_{ijL}\right)\left(1-\pi_{ijL}\right)\right| + \left|\left(\mu_{ijU} - \nu_{ijU}\right)\left(1-\pi_{ijU}\right)\right| \right)}{2},j=1,2,\cdots,n \tag{3.3} } Ej=m1i=1∑mcos2π(∣(μijL−νijL)(1−πijL)∣+∣(μijU−νijU)(1−πijU)∣),j=1,2,⋯,n(3.3)
于是可得属性 G j ( j = 1 , 2 , ⋯ , n ) G_{j}\left(j=1,2,\cdots,n\right) Gj(j=1,2,⋯,n)的权重 ω j \omega_{j} ωj
ω j = 1 − E j n − ∑ j = 1 n E j , j = 1 , 2 , ⋯ , n (3.4) \omega_{j} = \frac {1- E_{j}} {n - \sum_{j=1}^{n}E_{j}},j=1,2,\cdots,n \tag{3.4} ωj=n−∑j=1nEj1−Ej,j=1,2,⋯,n(3.4)
3.1.3 基于模糊熵的区间直觉模糊VIKOR方法的决策步骤
S.1 确定多属性决策问题的方案集 Y = { Y 1 , Y 2 , ⋯ , Y m } Y=\{Y_1,Y_2,\cdots,Y_m\} Y={Y1,Y2,⋯,Ym}和属性集 G = { G 1 , G 2 , ⋯ , G n } G=\{G_1,G_2,\cdots,G_n\} G={G1,G2,⋯,Gn},获取多属性决问题中方案 Y i ∈ Y Y_{i} \in Y Yi∈Y关于属性 G j ∈ G G_{j} \in G Gj∈G的区间直觉模糊特征信息,构建区间直觉模糊决策矩阵 F = ( ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ ) m × n F = \left(\left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle\right)_{m×n} F=(⟨[μijL,μijU],[νijL,νijU]⟩)m×n。
S.2 根据区间直觉模糊多属性决策矩阵 F F F,利用式 ( 3.3 ) (3.3) (3.3)和 ( 3.4 ) (3.4) (3.4)计算属性 G = { G 1 , G 2 , ⋯ , G n } G=\{G_1,G_2,\cdots,G_n\} G={G1,G2,⋯,Gn}的区间直觉模糊熵 E j E_{j} Ej和属性权重 ω j \omega_{j} ωj
S.3 根据区间直觉模糊多属性决策矩阵 F F F确定多属性决策问题的正理想解 Y + Y^{+} Y+和负理想解 Y − Y^{-} Y−:
Y + = ( Y 1 + , Y 2 + , ⋯ , Y n + ) = ( ⟨ [ μ 1 L + , μ 1 U + ] , [ ν 1 L + , ν 1 U + ] ⟩ , ⟨ [ μ 2 L + , μ 2 U + ] , [ ν 2 L + , ν 2 U + ] ⟩ , ⋯ , ⟨ [ μ n L + , μ n U + ] , [ ν n L + , ν n U + ] ⟩ ) \begin{aligned} Y^{+} &= \left( Y_{1}^{+},Y_{2}^{+},\cdots,Y_{n}^{+} \right) \\ &= \left(\left\langle \left[\mu_{1L}^{+},\mu_{1U}^{+}\right],\left[\nu_{1L}^{+},\nu_{1U}^{+}\right] \right\rangle,\left\langle\left[\mu_{2L}^{+},\mu_{2U}^{+}\right],\left[\nu_{2L}^{+},\nu_{2U}^{+}\right] \right\rangle,\cdots,\left\langle \left[\mu_{nL}^{+},\mu_{nU}^{+}\right],\left[\nu_{nL}^{+},\nu_{nU}^{+}\right] \right\rangle\right) \end{aligned} Y+=(Y1+,Y2+,⋯,Yn+)=(⟨[μ1L+,μ1U+],[ν1L+,ν1U+]⟩,⟨[μ2L+,μ2U+],[ν2L+,ν2U+]⟩,⋯,⟨[μnL+,μnU+],[νnL+,νnU+]⟩)
Y − = ( Y 1 − , Y 2 − , ⋯ , Y n − ) = ( ⟨ [ μ 1 L − , μ 1 U − ] , [ ν 1 L − , ν 1 U − ] ⟩ , ⟨ [ μ 2 L − , μ 2 U − ] , [ ν 2 L − , ν 2 U − ] ⟩ , ⋯ , ⟨ [ μ n L − , μ n U − ] , [ ν n L − , ν n U − ] ⟩ ) \begin{aligned} Y^{-} &= \left( Y_{1}^{-},Y_{2}^{-},\cdots,Y_{n}^{-} \right) \\ &= \left(\left\langle \left[\mu_{1L}^{-},\mu_{1U}^{-}\right],\left[\nu_{1L}^{-},\nu_{1U}^{-}\right] \right\rangle,\left\langle\left[\mu_{2L}^{-},\mu_{2U}^{-}\right],\left[\nu_{2L}^{-},\nu_{2U}^{-}\right] \right\rangle,\cdots,\left\langle \left[\mu_{nL}^{-},\mu_{nU}^{-}\right],\left[\nu_{nL}^{-},\nu_{nU}^{-}\right] \right\rangle\right) \end{aligned} Y−=(Y1−,Y2−,⋯,Yn−)=(⟨[μ1L−,μ1U−],[ν1L−,ν1U−]⟩,⟨[μ2L−,μ2U−],[ν2L−,ν2U−]⟩,⋯,⟨[μnL−,μnU−],[νnL−,νnU−]⟩)
式中
⟨ [ μ j L + , μ j U + ] , [ ν j L + , ν j U + ] ⟩ = ⟨ [ max i μ i j L , max i μ i j U ] , [ min i ν i j L , min i ν i j U ] ⟩ \left\langle \left[\mu_{jL}^{+},\mu_{jU}^{+}\right],\left[\nu_{jL}^{+},\nu_{jU}^{+}\right] \right\rangle = \left\langle \left[\max_{i}\mu_{ijL},\max_{i}\mu_{ijU}\right],\left[\min_{i}\nu_{ijL},\min_{i}\nu_{ijU}\right] \right\rangle ⟨[μjL+,μjU+],[νjL+,νjU+]⟩=⟨[imaxμijL,imaxμijU],[iminνijL,iminνijU]⟩
⟨ [ μ j L − , μ j U − ] , [ ν j L − , ν j U − ] ⟩ = ⟨ [ min i μ i j L , min i μ i j U ] , [ max i ν i j L , max i ν i j U ] ⟩ \left\langle \left[\mu_{jL}^{-},\mu_{jU}^{-}\right],\left[\nu_{jL}^{-},\nu_{jU}^{-}\right] \right\rangle = \left\langle \left[\min_{i}\mu_{ijL},\min_{i}\mu_{ijU}\right],\left[\max_{i}\nu_{ijL},\max_{i}\nu_{ijU}\right] \right\rangle ⟨[μjL−,μjU−],[νjL−,νjU−]⟩=⟨[iminμijL,iminμijU],[imaxνijL,imaxνijU]⟩
S.4 计算的群体效益值 S i S_i Si、个体遗憾值 R i R_i Ri:
S i = ∑ j = 1 n ω j [ d ( Y j + , F ~ i j ) d ( Y j + , Y j − ) ] , i = 1 , 2 , ⋯ , m S_{i} = \sum_{j=1}^{n}\omega_{j}\left[\frac{d\left(Y_{j}^{+},\tilde{F}_{ij}\right)}{d\left(Y_{j}^{+},Y_{j}^{-}\right)}\right],i=1,2,\cdots,m Si=j=1∑nωj⎣⎡d(Yj+,Yj−)d(Yj+,F~ij)⎦⎤,i=1,2,⋯,m
R i = max j { ω j [ d ( Y j + , F ~ i j ) d ( Y j + , Y j − ) ] } , i = 1 , 2 , ⋯ , m R_{i} = \max_{j} \left\{ \omega_{j}\left[ \frac{d\left(Y_{j}^{+},\tilde{F}_{ij}\right)}{d\left(Y_{j}^{+},Y_{j}^{-}\right)} \right] \right\},i=1,2,\cdots,m Ri=jmax⎩⎨⎧ωj⎣⎡d(Yj+,Yj−)d(Yj+,F~ij)⎦⎤⎭⎬⎫,i=1,2,⋯,m
式中, d ( Y j + , F ~ i j ) d\left(Y_{j}^{+},\tilde{F}_{ij}\right) d(Yj+,F~ij), d ( Y j + , Y j − ) d\left(Y_{j}^{+},Y_{j}^{-}\right) d(Yj+,Yj−)为两个区间直觉模糊数的距离,用以下公式计算:
d ( α ~ 1 , α ~ 2 ) = 1 2 ( ∣ μ α ~ 1 − μ α ~ 2 ∣ + ∣ ν α ~ 1 − ν α ~ 2 ∣ + ∣ π α ~ 1 − π α ~ 2 ∣ ) d\left(\tilde{\alpha}_{1}, \tilde{\alpha}_{2}\right)=\frac{1}{2}\left(\left|\mu_{\tilde{\alpha}_{1}}-\mu_{\tilde{\alpha}_{2}}\right|+\left|\nu_{\tilde{\alpha}_{1}}-\nu_{\tilde{\alpha}_{2}}\right|+\left|\pi_{\tilde{\alpha}_{1}}-\pi_{\tilde{\alpha}_{2}}\right|\right) d(α~1,α~2)=21(∣μα~1−μα~2∣+∣να~1−να~2∣+∣πα~1−πα~2∣)
其中, α ~ 1 = ⟨ μ α ~ 1 , ν α ~ 1 ⟩ \tilde{\alpha}_{1} = \left\langle \mu_{\tilde{\alpha}_{1}},\nu_{\tilde{\alpha}_{1}} \right\rangle α~1=⟨μα~1,να~1⟩, α ~ 2 = ⟨ μ α ~ 2 , ν α ~ 2 ⟩ \tilde{\alpha}_{2} = \left\langle \mu_{\tilde{\alpha}_{2}},\nu_{\tilde{\alpha}_{2}} \right\rangle α~2=⟨μα~2,να~2⟩为两个任意的区间直觉模糊数。
S.5 计算各备选方案 Y i ∈ Y Y_{i} \in Y Yi∈Y的折中值 Q i Q_{i} Qi:
Q i = ν S i − S + S − − S + + ( 1 − ν ) R i − R + R − − R + , i = 1 , 2 , ⋯ , m Q_{i} = \nu\frac{S_{i}-S^{+}}{S^{-}-S^{+}} + \left(1-\nu\right)\frac{R_{i}-R^{+}}{R^{-}-R^{+}},i=1,2,\cdots,m Qi=νS−−S+Si−S++(1−ν)R−−R+Ri−R+,i=1,2,⋯,m
其中, S i S_{i} Si为最大群体效用,是 L 1 , j L_{1,j} L1,j测度; R i R_{i} Ri取小个体遗憾,是 L ∞ , j L_{\infty,j} L∞,j测度; ω j \omega_{j} ωj为各属性权重; S + = min i ( S i ) S^{+}=\min_{i}(S_{i}) S+=mini(Si), S − = max i ( S i ) S^{-}=\max_{i}(S_{i}) S−=maxi(Si), R + = min i ( R i ) R^{+}=\min_{i}(R_{i}) R+=mini(Ri), R − = max i ( R i ) R^{-}=\max_{i}(R_{i}) R−=maxi(Ri); ν \nu ν为决策机制系数, ν ∈ [ 0 , 1 ] \nu \in [0,1] ν∈[0,1]。当 ν > 0.5 \nu \gt 0.5 ν>0.5时,表示根据最大群体效用的决策机制进行决策;当 ν = 0.5 \nu = 0.5 ν=0.5时,表示依据决策者经过协商达成共识的决策机制进行决策;当 ν < 0.5 \nu \lt 0.5 ν<0.5时,表示根据最小个体遗憾的决策机制进行决策。
S.6 设按 S i S_i Si、 R i R_i Ri、 Q i Q_i Qi值分别从小到大排序,确定妥协解方案。按 Q i Q_i Qi值递增得到的排序为 Y ( 1 ) , Y ( 2 ) , ⋯ , Y ( J ) , ⋯ , Y ( m ) Y^{(1)},Y^{(2)},\cdots,Y^{(J)},\cdots, Y^{(m)} Y(1),Y(2),⋯,Y(J),⋯,Y(m),则依据排序条件 1 1 1和排序条件 2 2 2可确定备选方案的优劣排序和最优方案或折中方案。
排序条件1 可接受优势条件: Q ( Y ( 2 ) ) − Q ( Y ( 1 ) ) ≥ 1 m − 1 Q\left(Y^{(2)}\right) - Q\left(Y^{(1)}\right) \geq \frac{1}{m-1} Q(Y(2))−Q(Y(1))≥m−11
排序条件2 决策过程中可接受的稳定性条件:方案 Y ( 1 ) Y^{(1)} Y(1)必须也是按照 S i S_i Si值或 R i R_i Ri值排序第一的方案。
3.2 属性权重信息未知情形下的区间直觉模糊VIKOR方法
3.2.1 问题描述
设某多属性决策问题有 m m m个方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m),组成方案集 Y = ( Y 1 , Y 2 , ⋯ , Y m ) Y=\left(Y_{1},Y_{2},\cdots,Y_{m}\right) Y=(Y1,Y2,⋯,Ym),评价每个方案的属性(或指标)为 G j ( j = 1 , 2 , ⋯ , n ) G_{j}\left(j=1,2,\cdots,n\right) Gj(j=1,2,⋯,n),记属性集为 G = { G 1 , G 2 , ⋯ , G n } G=\left\{ G_{1},G_{2},\cdots,G_{n} \right\} G={G1,G2,⋯,Gn},假设属性权重向量 ω = ( ω 1 , ω 2 , ⋯ , ω n ) T \omega = \left(\omega_{1},\omega{2},\cdots,\omega_{n}\right)^{T} ω=(ω1,ω2,⋯,ωn)T。如果 F ~ i j = ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ ( i = 1 , 2 , ⋯ , m ; j = 1 , 2 , ⋯ , n ) \tilde{F}_{ij} = \left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle \left(i=1,2,\cdots,m;j=1,2,\cdots,n\right) F~ij=⟨[μijL,μijU],[νijL,νijU]⟩(i=1,2,⋯,m;j=1,2,⋯,n)为区间直觉模糊集,表示方案满足属性 G j ∈ G G_{j} \in G Gj∈G和不满足属性 G j ∈ G G_{j} \in G Gj∈G的程度,且 0 ≤ μ i j + ν i j ≤ 1 0 \leq \mu_{ij}+\nu_{ij} \leq 1 0≤μij+νij≤1,矩阵 F = ( ⟨ [ μ i j L , μ i j U ] , [ ν i j L , ν i j U ] ⟩ ) m × n F = \left(\left\langle \left[\mu_{ijL},\mu_{ijU}\right], \left[\nu_{ijL},\nu_{ijU}\right] \right\rangle\right)_{m×n} F=(⟨[μijL,μijU],[νijL,νijU]⟩)m×n为该多属性决策问题的区间直觉模糊决策矩阵。现在的问题是依据区间直觉模糊决策矩阵 F F F,如何得到一个有效的决策分析方法来对所有方案进行优劣排序。
3.2.2 属性权重的确定方法
根据区间直觉模糊多属性决策矩阵 F F F确定多属性决策问题的正理想解 Y + Y^{+} Y+和负理想解 Y − Y^{-} Y−:
Y + = ( Y 1 + , Y 2 + , ⋯ , Y n + ) = ( ⟨ [ μ 1 L + , μ 1 U + ] , [ ν 1 L + , ν 1 U + ] ⟩ , ⟨ [ μ 2 L + , μ 2 U + ] , [ ν 2 L + , ν 2 U + ] ⟩ , ⋯ , ⟨ [ μ n L + , μ n U + ] , [ ν n L + , ν n U + ] ⟩ ) \begin{aligned} Y^{+} &= \left( Y_{1}^{+},Y_{2}^{+},\cdots,Y_{n}^{+} \right) \\ &= \left(\left\langle \left[\mu_{1L}^{+},\mu_{1U}^{+}\right],\left[\nu_{1L}^{+},\nu_{1U}^{+}\right] \right\rangle,\left\langle\left[\mu_{2L}^{+},\mu_{2U}^{+}\right],\left[\nu_{2L}^{+},\nu_{2U}^{+}\right] \right\rangle,\cdots,\left\langle \left[\mu_{nL}^{+},\mu_{nU}^{+}\right],\left[\nu_{nL}^{+},\nu_{nU}^{+}\right] \right\rangle\right) \end{aligned} Y+=(Y1+,Y2+,⋯,Yn+)=(⟨[μ1L+,μ1U+],[ν1L+,ν1U+]⟩,⟨[μ2L+,μ2U+],[ν2L+,ν2U+]⟩,⋯,⟨[μnL+,μnU+],[νnL+,νnU+]⟩)
Y − = ( Y 1 − , Y 2 − , ⋯ , Y n − ) = ( ⟨ [ μ 1 L − , μ 1 U − ] , [ ν 1 L − , ν 1 U − ] ⟩ , ⟨ [ μ 2 L − , μ 2 U − ] , [ ν 2 L − , ν 2 U − ] ⟩ , ⋯ , ⟨ [ μ n L − , μ n U − ] , [ ν n L − , ν n U − ] ⟩ ) \begin{aligned} Y^{-} &= \left( Y_{1}^{-},Y_{2}^{-},\cdots,Y_{n}^{-} \right) \\ &= \left(\left\langle \left[\mu_{1L}^{-},\mu_{1U}^{-}\right],\left[\nu_{1L}^{-},\nu_{1U}^{-}\right] \right\rangle,\left\langle\left[\mu_{2L}^{-},\mu_{2U}^{-}\right],\left[\nu_{2L}^{-},\nu_{2U}^{-}\right] \right\rangle,\cdots,\left\langle \left[\mu_{nL}^{-},\mu_{nU}^{-}\right],\left[\nu_{nL}^{-},\nu_{nU}^{-}\right] \right\rangle\right) \end{aligned} Y−=(Y1−,Y2−,⋯,Yn−)=(⟨[μ1L−,μ1U−],[ν1L−,ν1U−]⟩,⟨[μ2L−,μ2U−],[ν2L−,ν2U−]⟩,⋯,⟨[μnL−,μnU−],[νnL−,νnU−]⟩)
式中
⟨ [ μ j L + , μ j U + ] , [ ν j L + , ν j U + ] ⟩ = ⟨ [ max i μ i j L , max i μ i j U ] , [ min i ν i j L , min i ν i j U ] ⟩ \left\langle \left[\mu_{jL}^{+},\mu_{jU}^{+}\right],\left[\nu_{jL}^{+},\nu_{jU}^{+}\right] \right\rangle = \left\langle \left[\max_{i}\mu_{ijL},\max_{i}\mu_{ijU}\right],\left[\min_{i}\nu_{ijL},\min_{i}\nu_{ijU}\right] \right\rangle ⟨[μjL+,μjU+],[νjL+,νjU+]⟩=⟨[imaxμijL,imaxμijU],[iminνijL,iminνijU]⟩
⟨ [ μ j L − , μ j U − ] , [ ν j L − , ν j U − ] ⟩ = ⟨ [ min i μ i j L , min i μ i j U ] , [ max i ν i j L , max i ν i j U ] ⟩ \left\langle \left[\mu_{jL}^{-},\mu_{jU}^{-}\right],\left[\nu_{jL}^{-},\nu_{jU}^{-}\right] \right\rangle = \left\langle \left[\min_{i}\mu_{ijL},\min_{i}\mu_{ijU}\right],\left[\max_{i}\nu_{ijL},\max_{i}\nu_{ijU}\right] \right\rangle ⟨[μjL−,μjU−],[νjL−,νjU−]⟩=⟨[iminμijL,iminμijU],[imaxνijL,imaxνijU]⟩
对于决策方案方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m),用 d + ( ω ) d^{+}(\omega) d+(ω)和 d − ( ω ) d^{-}(\omega) d−(ω)分别表示方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)到正理想解 Y + Y^{+} Y+和负理想解 Y − Y^{-} Y−的加权偏差,定义 d + ( ω ) d^{+}(\omega) d+(ω)和 d − ( ω ) d^{-}(\omega) d−(ω)分别为:
d i + ( ω ) = 1 4 ∑ j = 1 n ω j [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] \color{blue} { d_{i}^{+}(\omega) = \frac{1}{4} \sum_{j=1}^{n}\omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right] } di+(ω)=41j=1∑nωj[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣]
d i − ( ω ) = 1 4 ∑ j = 1 n ω j [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] \color{blue} { d_{i}^{-}(\omega) = \frac{1}{4} \sum_{j=1}^{n}\omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right] } di−(ω)=41j=1∑nωj[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣]
从对决策方案进行排序角度考虑,若方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)离正理想解越近、离负理想解越远,则方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)越优。对于给定的权重向量 ω = ( ω 1 , ω 2 , ⋯ , ω n ) T \omega = \left(\omega_{1},\omega{2},\cdots,\omega_{n}\right)^{T} ω=(ω1,ω2,⋯,ωn)T, d + ( ω ) d^{+}(\omega) d+(ω)越小而 d − ( ω ) d^{-}(\omega) d−(ω)越大,则相应的方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)越优。
当属性权重信息不完全时,若权重向量满足: ω ‾ j ≤ ω j ≤ ω ‾ j ∑ j = 1 n ω j 2 = 1 , ω j ≥ 0 ( j = 1 , 2 , ⋯ , n ) {\underline{\omega}_{j} \leq \omega_{j} \leq {\overline{\omega}_{j}}} \sum_{j=1}^{n}\omega_{j}^{2} = 1,\omega_{j} \geq 0 \left(j=1,2,\cdots,n\right) ωj≤ωj≤ωj∑j=1nωj2=1,ωj≥0(j=1,2,⋯,n),可建立多目标优化模型:
{ min d i + ( ω ) = 1 4 ∑ j = 1 n ω j [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] , i = 1 , 2 , ⋯ , m max d i − ( ω ) = 1 4 ∑ j = 1 n ω j [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] , i = 1 , 2 , ⋯ , m s.t. ω ‾ j ⩽ ω j ⩽ ω ˉ j , ∑ j = 1 n ω j = 1 , ω j ⩾ 0 , j = 1 , 2 , ⋯ , n (3.5) \left\{ \begin{array}{l} \min d_{i}^{+}(\omega)=\frac{1}{4} \sum_{j=1}^{n} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right], i=1,2, \cdots, m \\ \max d_{i}^{-}(\omega)=\frac{1}{4} \sum_{j=1}^{n} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right], i=1,2, \cdots, m \\ \text { s.t. } \underline{\omega}_{j} \leqslant \omega_{j} \leqslant \bar{\omega}_{j}, \sum_{j=1}^{n} \omega_{j}=1, \omega_{j} \geqslant 0, j=1,2, \cdots, n\end{array} \right. \tag{3.5} ⎩⎪⎪⎨⎪⎪⎧mindi+(ω)=41∑j=1nωj[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣],i=1,2,⋯,mmaxdi−(ω)=41∑j=1nωj[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣],i=1,2,⋯,m s.t. ωj⩽ωj⩽ωˉj,∑j=1nωj=1,ωj⩾0,j=1,2,⋯,n(3.5)
令
d + ( ω ) = ∑ i = 1 m d i + ( ω ) = 1 4 ∑ j = 1 n ∑ i = 1 m ω j [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] \begin{aligned} d^{+}(\omega) &= \sum_{i=1}^{m} d_{i}^{+}(\omega) \\ &= \frac{1}{4} \sum_{j=1}^{n} \sum_{i=1}^{m} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right] \end{aligned} d+(ω)=i=1∑mdi+(ω)=41j=1∑ni=1∑mωj[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣]
d − ( ω ) = ∑ i = 1 m d i − ( ω ) = 1 2 ∑ j = 1 n ∑ i = 1 m ω j [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] \begin{aligned} d^{-}(\omega) &= \sum_{i=1}^{m} d_{i}^{-}(\omega) \\ &= \frac{1}{2} \sum_{j=1}^{n} \sum_{i=1}^{m} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right] \end{aligned} d−(ω)=i=1∑mdi−(ω)=21j=1∑ni=1∑mωj[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣]
则 d + ( ω ) d^{+}(\omega) d+(ω)、 d − ( ω ) d^{-}(\omega) d−(ω)分别表示所有方案 Y i ( i = 1 , 2 , ⋯ , m ) Y_{i}\left(i=1,2,\cdots,m\right) Yi(i=1,2,⋯,m)到正理想解 Y + Y^{+} Y+和负理想解 Y − Y^{-} Y−的加权偏差总和。每个方案都是公平竞争的,不存在任何偏好关系,因此模型 ( 2.10 ) (2.10) (2.10)等权集结为如下单目标最优化模型:
{ min d + ( ω ) − d − ( ω ) = 1 4 ∑ j = 1 n ∑ i = 1 m ω j [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ − ∣ μ i j L − μ j L − ∣ − ∣ μ i j U − μ j U − ∣ − ∣ ν i j L − ν j L − ∣ − ∣ ν i j U − ν j U − ∣ ] s.t. ω ‾ j ⩽ ω j ⩽ ω ˉ j , ∑ j = 1 n ω j = 1 , ω j ⩾ 0 , j = 1 , 2 , ⋯ , n (3.6) \left\{ \begin{array}{c} \min d^{+}(\omega)-d^{-}(\omega)=\frac{1}{4} \sum_{j=1}^{n} \sum_{i=1}^{m} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right. \\ \left.\quad - \left|\mu_{ijL}-\mu_{jL}^{-}\right| - \left|\mu_{ijU}-\mu_{jU}^{-}\right| - \left|\nu_{ijL}-\nu_{jL}^{-}\right| - \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right] \\ \text {s.t. } \underline{\omega}_{j} \leqslant \omega_{j} \leqslant \bar{\omega}_{j}, \sum_{j=1}^{n} \omega_{j}=1, \omega_{j} \geqslant 0, j=1,2, \cdots, n \end{array} \right. \tag{3.6} ⎩⎪⎪⎨⎪⎪⎧mind+(ω)−d−(ω)=41∑j=1n∑i=1mωj[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣−∣∣∣μijL−μjL−∣∣∣−∣∣∣μijU−μjU−∣∣∣−∣∣∣νijL−νjL−∣∣∣−∣∣∣νijU−νjU−∣∣∣]s.t. ωj⩽ωj⩽ωˉj,∑j=1nωj=1,ωj⩾0,j=1,2,⋯,n(3.6)
解最优化模型 ( 3.6 ) (3.6) (3.6),可得属性权重向量 ω = ( ω 1 , ω 2 , ⋯ , ω n ) T \omega = \left(\omega_{1},\omega{2},\cdots,\omega_{n}\right)^{T} ω=(ω1,ω2,⋯,ωn)T。
如果属性权重信息完全未知,则可构建以 d + ( ω ) d^{+}(\omega) d+(ω)最小化或 d − ( ω ) d^{-}(\omega) d−(ω)最大化为目标的最优化模型式 ( 3.7 ) (3.7) (3.7)和式 ( 3.8 ) (3.8) (3.8):
{ min d + ( ω ) = 1 4 ∑ j = 1 n ∑ i = 1 m ω j [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] s.t. ∑ j = 1 n ω j 2 = 1 , ω j ⩾ 0 , j = 1 , 2 , ⋯ , n (3.7) \left\{ \begin{array}{l} \min d^{+}(\omega)=\frac{1}{4} \sum_{j=1}^{n} \sum_{i=1}^{m} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right] \\ \text {s.t. } \sum_{j=1}^{n} \omega_{j}^{2}=1, \omega_{j} \geqslant 0, j=1,2, \cdots, n \end{array} \right. \tag{3.7} {mind+(ω)=41∑j=1n∑i=1mωj[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣]s.t. ∑j=1nωj2=1,ωj⩾0,j=1,2,⋯,n(3.7)
{ max d + ( ω ) = 1 2 ∑ j = 1 n ∑ i = 1 m ω j [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] s.t. ∑ j = 1 n ω j 2 = 1 , ω j ⩾ 0 , j = 1 , 2 , ⋯ , n (3.8) \left\{ \begin{array}{l} \max d^{+}(\omega)=\frac{1}{2} \sum_{j=1}^{n} \sum_{i=1}^{m} \omega_{j}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right] \\ \text {s.t. } \sum_{j=1}^{n} \omega_{j}^{2}=1, \omega_{j} \geqslant 0, j=1,2, \cdots, n \end{array} \right. \tag{3.8} {maxd+(ω)=21∑j=1n∑i=1mωj[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣]s.t. ∑j=1nωj2=1,ωj⩾0,j=1,2,⋯,n(3.8)
通过构造拉格朗日函数,利用极值理论求解最优化模型式 ( 3.7 ) (3.7) (3.7)或式 ( 3.8 ) (3.8) (3.8),可得其最优解分别为
ω j ∗ = ∑ i = 1 m [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] ∑ j = 1 n [ ∑ i = 1 m [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] ] 2 , j = 1 , 2 , ⋯ , n (3.9) \omega_{j}^{*}=\frac{\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right]}{\sqrt{\sum_{j=1}^{n}\left[\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right] \right]^{2}}}, \quad j=1,2, \cdots, n \tag{3.9} ωj∗=∑j=1n[∑i=1m[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣]]2∑i=1m[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣],j=1,2,⋯,n(3.9)
或
ω j ∗ = ∑ i = 1 m [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] ∑ j = 1 n [ ∑ i = 1 m [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] ] 2 , j = 1 , 2 , ⋯ , n (3.10) \omega_{j}^{*}=\frac{\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right]}{\sqrt{\sum_{j=1}^{n}\left[\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right]\right]^{2}}}, \quad j=1,2, \cdots, n \tag{3.10} ωj∗=∑j=1n[∑i=1m[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣]]2∑i=1m[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣],j=1,2,⋯,n(3.10)
对 ω j ∗ \omega_{j}^{*} ωj∗进行归一化处理可得属性 G j ( j = 1 , 2 , ⋯ , n ) G_{j}\left(j=1,2,\cdots,n\right) Gj(j=1,2,⋯,n)的权重为:
ω j = ∑ i = 1 m [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] ∑ j = 1 n ∑ i = 1 m [ ∣ μ i j L − μ j L + ∣ + ∣ μ i j U − μ j U + ∣ + ∣ ν i j L − ν j L + ∣ + ∣ ν i j U − ν j U + ∣ ] , j = 1 , 2 , ⋯ , n (3.11) \omega_{j} = \frac {\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right]} {\sum_{j=1}^{n}\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{+}\right| + \left|\mu_{ijU}-\mu_{jU}^{+}\right| + \left|\nu_{ijL}-\nu_{jL}^{+}\right| + \left|\nu_{ijU}-\nu_{jU}^{+}\right| \right]}, \quad j=1,2, \cdots, n \tag{3.11} ωj=∑j=1n∑i=1m[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣]∑i=1m[∣∣∣μijL−μjL+∣∣∣+∣∣∣μijU−μjU+∣∣∣+∣∣∣νijL−νjL+∣∣∣+∣∣∣νijU−νjU+∣∣∣],j=1,2,⋯,n(3.11)
或
ω j = ∑ i = 1 m [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] ∑ j = 1 n ∑ i = 1 m [ ∣ μ i j L − μ j L − ∣ + ∣ μ i j U − μ j U − ∣ + ∣ ν i j L − ν j L − ∣ + ∣ ν i j U − ν j U − ∣ ] , j = 1 , 2 , ⋯ , n (3.12) \omega_{j} = \frac {\sum_{i=1}^{m}\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right]} {\sum_{j=1}^{n}\sum_{i=1}^{m} {\left[ \left|\mu_{ijL}-\mu_{jL}^{-}\right| + \left|\mu_{ijU}-\mu_{jU}^{-}\right| + \left|\nu_{ijL}-\nu_{jL}^{-}\right| + \left|\nu_{ijU}-\nu_{jU}^{-}\right| \right]}}, \quad j=1,2, \cdots, n \tag{3.12} ωj=∑j=1n∑i=1m[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣]∑i=1m[∣∣∣μijL−μjL−∣∣∣+∣∣∣μijU−μjU−∣∣∣+∣∣∣νijL−νjL−∣∣∣+∣∣∣νijU−νjU−∣∣∣],j=1,2,⋯,n(3.12)