CN105117820A - Grain storage green degree evaluating method based on DEA-AHP - Google Patents

Abstract

The invention provides a grain storage green degree evaluating method based on DEA-AHP, which belongs to the grain storage technical field. The method comprises the following steps of (1) determining an evaluating index system by means of an AHP (analytic hierarchy process) method, making the relations of each factor hierarchical and principled and distinguishing the affecting degree of each factor on an evaluation object; (2) determining the specific value of each index correspondingly in each grain storage scheme by using a DEA (data envelope approach); (2) and comprehensively evaluating the grain storage schemes in a way to consider the index weights. The green degree of each grain storage factor can be investigated. Therefore, the advantages and disadvantages of each grain storage factor are found. Further improvement can be facilitated. Meanwhile, multiple grain storage schemes are discussed and calculated together. Relatively strong comparability is achieved. High evaluating efficiency is further achieved.

Description

A kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP

Technical field

The invention belongs to View of Grain Storage Technoiogy field, be specifically related to a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP.

Background technology

Under national policy is vigorously supported, China's grain-production continuous 8 years high yields, but demand is also ever-increasing while, supply and demand is still tight slightly.Just when this supply and demand is difficult to change, more highlight the importance of foodstuff preservation work.Under " grain security " intension of second time World Food Summit in 1996 proposition requires, ensure the level of reserve of grain, to go mouldy from annual grain and the loss such as insect pest (loss is about 8% of grain yield) is started with, will Loss reducing as much as possible, ensureing nutrition and the health of grain qualitatively, carry out foodstuff preservation work.These are all for the new demand of View of Grain Storage Technoiogy work, new challenge, are also the Missions that history is given.

The direction of China's foodstuff preservation development, mainly upgraded by traditional grain storage technology, realize " green, ecological, harmonious " of View of Grain Storage Technoiogy, finally realize the target of foodstuff preservation " high-quality, high nutrition, high benefit " and " low-loss, low stain, low cost ".With this most important theories of ecological grain storage for instructing, suit measures to local conditions to set up China's characteristic, different Grain Reserve Ecology district, optimal economic operational mode Green Grain Storage technical system, from the Natural resources condition utilizing various ecologic environment to greatest extent, to reach the final purpose of safe grain storage.Therefore, the greenness of foodstuff preservation has become the important indicator weighing the advanced degree of China's View of Grain Storage Technoiogy.

Current China not yet forms the foodstuff preservation green degree evaluation system of system, as the aspects such as in GB/T29890-2013 the storage facilities to grain depot and equipment, harmful organism control propose some basic demands, paper " the current status and prospect trend of China's foodstuff preservation " (Yang Guangjing, Ren Yunhong, Jia Jinyuan, Deng. the current status and prospect of China's foodstuff preservation becomes, grain processing, 2012,37 (1): 60-63.) set forth the aspects such as the present situation of China's foodstuff preservation, technical standard specification, development trend, but its appraisement system has not been introduced, paper " Evaluation System On Safe Grain-storage Technological Indexes " (Tao Cheng, Wu Xia, Xu Hongsheng, Deng. Evaluation System On Safe Grain-storage Technological Indexes. foodstuff preservation, 2004, 32 (5): 15-21.) the storage facilities configuration that national each Grain Reserve Ecology region is different is comprehensively analyzed, grain quality and grain heap ecological condition are on the impact of Grain Security Situation, but its subjective estimate method adopting expertise to judge, evaluation result is easily by the experience of expert, the impact of level and artificial subjective factor, it is objective not and accurate to evaluate, and evaluation index arranges comparatively single, lack harmonious, inadequate science, therefore urgent need finds a kind of by objective for the subjectivity evaluation method combined.

Analytical hierarchy process (Analytichierarchyprocess, AHP) is taught by the scholar SAATY that plans strategies for of famous American that a kind of quantitative and qualitative analysis proposed 20 century 70 mid-terms combines, the analytical approach of systematized, stratification.Although but traditional analytical hierarchy process can make full use of the subjective suggestion of expert, it too relies on subjective judgement, its Scale Method is Deterministic Methods, therefore has obvious deficiency.

DEA (DEA) is based on " relative efficiency " concept, drops into a kind of systematic analytic method carrying out relative effectiveness or benefit evaluation with the unit (department) of multi objective output to identical type according to multi objective.Normally to one group of given decision package (DMU), by the comprehensive analysis of input and output data, DEA can draw the quantitative index of each DMU overall efficiency.The evaluation result of DEA Method is not affected by human factors, but can not reflect the preference of decision maker.

In sum, foodstuff preservation greenness is the green degree of foodstuff preservation, and the greenization of foodstuff preservation refers to the harm to grain such as Antifungi, insect pest, humidity in foodstuff preservation process, foodstuff preservation advanced technology and equipment are applied fully, reaches the unification of economic benefit, social benefit and environmental benefit.The dry technology, ca cold storage technology, mechanical ventilator, grain depot layout, grain feelings detection technique etc. of grain depot all have larger impact to the greenness of grain depot, use the greenness reliability of single evaluation method evaluation foodstuff preservation lower when multiple conditional decision.Current China not yet forms the foodstuff preservation green degree evaluation system of system, utilizes step analysis not have especially in conjunction with the foodstuff preservation Green Degree Evaluation of data envelopment.

Summary of the invention

The object of the invention is to the deficiency overcoming foodstuff preservation greenness Enterprises ' Green Degree degree method, invent a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP, propose 13 representative indexs, and be have rated the single factor greenness performance of 13 indexs in conjunction with DEA by AHP, and overcome the deficiency that simple layer fractional analysis subjectivity is strong, and the scheme of multiple foodstuff preservation can be put together and discuss and calculate, the method can not only reflect the preference of decision maker, and can the evaluation alternatives of objective quantification, a kind of next objective, science, quantifiable foodstuff preservation Green Degree Evaluation.

For reaching above object, the present invention adopts several lower technical scheme, comprises the following steps:

S1, first according to China's View of Grain Storage Technoiogy by traditional grain storage technology to Green Grain Storage technical development, the requirement developed to the warehousing management that becomes more meticulous by extensive warehousing management, the present invention proposes 13 representative indexs, then analytical hierarchy process (Analytichierarchyprocess is used, AHP) assessment indicator system is determined, make the relational hierarchy between each factor, methodization, and the influence degree of each factor to evaluation objective can be distinguished.Use AHP method, organize the relative importance of expert's Comparative indices between two, provide corresponding proportional roles, build upper strata key element to the judgment matrix of lower floor's coherent element, and calculate eigenvalue of maximum and the character pair vector of judgment matrix, consistency check is passed through to the judgment matrix obtained.

S2, then determine the concrete manifestation of the corresponding each index of each foodstuff preservation scheme in conjunction with data envelope analysis (DEA).

S3, finally in conjunction with index weights, comprehensive evaluation is carried out to multiple foodstuff preservation scheme.

Further, described step S1, it comprises:

S11, agriculture products system refer to: according to China's View of Grain Storage Technoiogy by traditional grain storage technology to Green Grain Storage technical development, the feature developed to the warehousing management that becomes more meticulous by extensive warehousing management, the present invention proposes from green storage advanced technology A1, and green storage equipment and device A2 and green are stored in a warehouse and considered these 3 angles (first class index) of A3 and select 13 representative indexs (two-level index) as the foundation evaluating foodstuff preservation greenness.

Wherein green storage advanced technology A1 comprises: non-chemically Pesticidal technology B1, fungus control technology B2, advanced dry technology B3, air-conditioning storing grain technology B4 and low temperature treatment technology B5;

Green storage equipment and device A2 comprise: recirculation fumigation equipment B6, paddy drying equipment B 7, mechanical ventilator B8 and grain condition monitoring device B9;

Green storage considers A3 and comprises: rationally B13 safeguarded by grain depot addressing B10, rationally grain depot layout B11, Rational Matching facility B12 and reasonable grain depot.

S12, the judgment matrix of structure upper strata key element to lower floor's coherent element refer to: according to the comparison scale development of judgment matrix D of hierarchy Model according to the form below 1, respectively to organizing element development of judgment matrix between interior and group, namely respectively with first class index A1, A2, A3 for criterion, compare between two between group, obtain a judgment matrix D ₁; Then in group, respectively with each index for criterion, compare between two, judge and the relation of other indexs, obtain judgment matrix D ₂, D ₃, D ₄, 4 matrixes can be obtained altogether.The comparison scale of each index is as shown in table 1, and wherein i and j represents i-th key element and a jth key element in table respectively, d _ijrefer to the affinity criterions value of i-th key element and a jth key element.

The comparison scale of each index of table 1.

Judgment matrix D according to obtaining with co-relation below:

$D=\left[\begin{array}{cccc}{d}_{11}& d_{12}& ...& d_{1n}\\ d_{21}& d_{22}& ...& d_{2n}\\ \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}\\ ...\\ \end{array}& \begin{array}{c}.\\ .\\ .\end{array}\\ d_{n1}& d_{n2}& ...& d_{nn}\end{array}\right]$

S13, the eigenvalue of maximum calculating judgment matrix and character pair vector refer to: the eigenwert of judgment matrix D can be calculated by D × X=λ × X, and wherein λ is the eigenwert of matrix D, and non-vanishing vector X is called the proper vector corresponding to eigenvalue λ of D.By the Maximum characteristic root λ of the known D existence anduniquess of mathematical theory _max, adopt root method to calculate λ _maxwith the approximate value of W.First product M is done to the value of the element of the every a line in judgment matrix D _i=d _i1× d _i2× ... × d _in, n is judgment matrix exponent number, M _irepresent i-th key element importance ratio that is relative and other key elements long-pending, then to M _iopen n power, obtain again to W _ibe normalized, namely w _ifor the weight vectors of each factor after normalization, the weight of all two-level index equals the weight of each two-level index in group and the product of its place first class index shared weight between group.Finally by publicity try to achieve the eigenvalue of maximum λ of judgment matrix _max, wherein D is judgment matrix, and w is the weight vectors of each factor after normalization, and n is the exponent number of matrix D.

S14, Consistency Check in Judgement Matrix refer to: the D that analyst sets up according to personal experience and knowledge exists error unavoidably, for making judged result and actual conditions coincide better, and can by formula C _r=C _i/ R _ithe consistance of D is tested, in formula: C _ifor consistency check index, C _i=(λ _max-n)/(n-1); N is the exponent number of judgment matrix; R _ifor Aver-age Random Consistency Index, value sees the following form.Work as C _r<0.1, then the consistance of D can accept, and the above-mentioned w calculated is the weight vectors of each factor; Otherwise D need be readjusted to meeting consistency check requirement.

Table 2.R _ivalue

Judgment matrix exponent number	1	2	3	4	5	6	7	8	9
										R I	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45

Further, described step S2, it comprises:

S21, data envelope analysis specific algorithm refer to: wherein V=(v ₁, v ₂..., v _m) be the input of DEA, m is input variable quantity; U=(u ₁, u ₂..., u _k) be the output of DEA, k is output variable quantity; With " poor " (v in the present invention ₁), " in " (v ₂) be the input of system, " good " (u ₁), " excellent " (u ₂) as the output of system, i.e. V=(v ₁, v ₂) _,u=(u ₁, u ₂).Note $X_j=\left(\begin{array}{c}{x}_{1j}\\ x_{2j}\\ ...\\ x_{mj}\end{array}\right);Y_j=\left(\begin{array}{c}{y}_{1j}\\ y_{2j}\\ ...\\ y_{kj}\end{array}\right),$ J=1,2 ..., 13; Wherein X _jfor the score matrix of system input, Y _jfor the score matrix that system exports, X in the present invention _jnamely refer to that grain depot number of votes obtained of " poor " on index j is x _1j, " in " number of votes obtained be x _2j, the number of votes obtained of " good " is y _1j, the number of votes obtained of " excellent " is y _2j, that is to say the evaluation result of expert, by following formula:

$\left\{\begin{array}{c}\frac{U\&CenterDot;Y}{V\&CenterDot;X}\&le;1\\ V\&CenterDot;X=1\\ U\&GreaterEqual;0\\ V\&GreaterEqual;0\end{array}\right.$

$\max X_{ij}=\frac{U\&CenterDot;Y}{V\&CenterDot;X}$

Wherein UY is u ₁y _1j+ u ₂y _2j+ ... u _ky _mj, VX is v ₁x _1j+ v ₂x _2j+ ... v _kx _mj, maxX _ijfor the performance of grain depot i on index j, finally obtain the performance of each grain depot in 13 indexs.

Further, described step S3, it comprises:

S31, carry out comprehensive evaluation in conjunction with index weights and refer to: the weight G of each grain depot in every evaluation index determined according to data envelope analysis _i={ maxX _i1, maxX _i2..., maxX _i13, wherein i here refers to grain depot i, maxX _ijrefer to i-th performance of grain depot in a jth index, in conjunction with the weight sets L={w of 13 indexs determined according to analytical hierarchy process _b1, w _b2..., w _b13, w _bibe the weight of i-th factor, calculate the comprehensive evaluation score of each grain depot foodstuffs storage greenness, by comprehensive evaluation score can be calculated.Wherein, A _ifor the comprehensive evaluation score of grain depot i foodstuff preservation greenness.

Beneficial effect of the present invention:

1, set up a kind of easy, science, quantifiable foodstuff preservation Green Degree Evaluation, fill up the blank that current China not yet forms the foodstuff preservation Green Degree Evaluation of system.

2, according to China's View of Grain Storage Technoiogy by traditional grain storage technology to Green Grain Storage technical development, the feature developed to the warehousing management that becomes more meticulous by extensive warehousing management, propose 13 representative indexs, and the single factor greenness performance of these 13 indexs has been investigated by analytical hierarchy process, as wherein advanced dry technology weight accounts for 0.2262, air-conditioning storing grain technology weight accounts for 0.1376, recirculation fumigation equipment weight accounts for 0.0540, compensate for and use the shortcoming that the greenness reliability of single evaluation method evaluation foodstuff preservation is lower when multiple conditional decision at present, and judged by consistance, make according to personal experience's knowledge set up judged result can better and actual conditions combination, make appraisement system science more.

3, in conjunction with data envelope analysis, comprehensive evaluation is carried out to the greenness of grain depot by analytical hierarchy process, overcome the deficiency that simple layer fractional analysis subjectivity is strong, and the scheme of multiple foodstuff preservation can be put together and discuss and calculate, the method can not only reflect the preference of decision maker, and can the evaluation alternatives of objective quantification, result is as comprehensive evaluation score A _first=0.6895, A _second=0.5762, A _third=0.6623, A _fourthshown in=0.9446, second < third < first < fourth, embody stronger comparability, assess effectiveness is high.In addition, also can investigate foodstuff preservation and show at the greenness of single factor, thus find its relative merits, as found " grain depot first is weaker in grain depot layout, and lacks coming into operation of recirculation fumigation equipment ", so that further Improvement and perfection.

Accompanying drawing explanation

Fig. 1 is the flow chart of steps of a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP of the present invention;

Fig. 2 is the foodstuff preservation greenness index system figure that the present invention determines according to analytical hierarchy process;

Embodiment

As shown in Figure 1, the invention provides a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP, comprise the following steps:

4 grain depots are chosen in the present invention, first, second, third, fourth (first: Rudong County's grain and oil purchase and sale company, second: Cha Heyou meter company limited of Rudong County, third: Dantu District, Zhengjiang City grain and oil purchase and sale main office, fourth: Center Grain Reserves Nanjing Direct Affiliated Depot) as experimental subjects.

S1, first by analytical hierarchy process determination assessment indicator system, make the relational hierarchy between each factor, methodization, and the influence degree of each factor to evaluation objective can be distinguished.Use AHP method, organize the relative importance of expert's Comparative indices between two, provide corresponding proportional roles, build upper strata key element to the judgment matrix of lower floor's coherent element, and calculate eigenvalue of maximum and the character pair vector of judgment matrix, consistency check is passed through to the judgment matrix obtained.

The present invention selects yaahpV6.0 software to adopt analytical hierarchy process to calculate.

S2, then data envelope analysis is utilized to determine the concrete manifestation of the corresponding each index of each foodstuff preservation scheme.

The present invention selects lingo11 software to adopt envelope method to calculate.

S3, finally in conjunction with index weights, comprehensive evaluation is carried out to multiple foodstuff preservation scheme.

Further, described step S1, it comprises:

S11, as shown in Figure 2, agriculture products system refers to: according to China's View of Grain Storage Technoiogy by traditional grain storage technology to Green Grain Storage technical development, the feature developed to the warehousing management that becomes more meticulous by extensive warehousing management.Propose from green storage advanced technology A1, green storage equipment and device A2 and green are stored in a warehouse and are considered these 3 angles (first class index) of A3 and select 13 representative indexs (two-level index) as the foundation evaluating foodstuff preservation greenness.Wherein green storage advanced technology A1 comprises: non-chemically Pesticidal technology B1, fungus control technology B2, advanced dry technology B3, air-conditioning storing grain technology B4 and low temperature treatment technology B5; Green storage equipment and device A2 comprise: recirculation fumigation equipment B6, paddy drying equipment B 7, mechanical ventilator B8 and grain condition monitoring device B9; Green storage considers A3 and comprises: rationally B13 safeguarded by grain depot addressing B10, rationally grain depot layout B11, Rational Matching facility B12 and reasonable grain depot.

S12, the judgment matrix of structure upper strata key element to lower floor's coherent element refer to: according to the comparison scale development of judgment matrix D of hierarchy Model according to the form below.

$D=\left[\begin{array}{cccc}{d}_{11}& d_{12}& ...& d_{1n}\\ d_{21}& d_{22}& ...& d_{2n}\\ \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}.\\ .\\ .\end{array}& \begin{array}{c}\\ ...\\ \end{array}& \begin{array}{c}.\\ .\\ .\end{array}\\ d_{n1}& d_{n2}& ...& d_{nn}\end{array}\right]$

As shown above, wherein i and j represents i-th key element and a jth key element in table to the comparison scale of each index respectively, d _ijrefer to the affinity criterions value of i-th key element and a jth key element.

S13, the eigenvalue of maximum calculating judgment matrix and character pair vector refer to: the eigenwert of judgment matrix D can be calculated by D × X=λ × X, and wherein λ is the eigenwert of matrix D, and non-vanishing vector X is called the proper vector corresponding to eigenvalue λ of D.By the Maximum characteristic root λ of the known D existence anduniquess of mathematical theory _max, adopt root method to calculate λ _maxwith the approximate value of W.First product M is done to the value of the element of the every a line in judgment matrix D _i=d _i1× d _i2× ... × d _in, n is judgment matrix exponent number, M _irepresent i-th key element importance ratio that is relative and other key elements long-pending, then to M _iopen n power, obtain again to W _ibe normalized, namely w _ifor the weight vectors of each factor after normalization, the weight of all two-level index equals the weight of each two-level index in group and the product of its place first class index shared weight between group.Finally by publicity try to achieve the eigenvalue of maximum λ of judgment matrix _max, wherein D is judgment matrix, and w is the weight vectors of each factor after normalization, and n is the exponent number of matrix D.

S14, Consistency Check in Judgement Matrix refer to: the D that analyst sets up according to personal experience and knowledge exists error unavoidably, for making judged result and actual conditions coincide better, and can by formula C _r=C _i/ R _ithe consistance of D is tested, in formula: C _ifor consistency check index, C _i=(λ _max-n)/(n-1); N is the exponent number of judgment matrix; R _ifor Aver-age Random Consistency Index, value sees the following form.Work as C _r<0.1, then the consistance of D can accept, otherwise need readjust D to meeting consistency check requirement.

Judgment matrix exponent number	1	2	3	4	5	6	7	8	9
										R I	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45

If D meets consistency check, then the above-mentioned w calculated is the weight vectors of each factor.

The discrimination matrix D of 3 first class index is obtained according to expertise ₁as follows with 13 two-level index discrimination matrix:

$D_1=\left[\begin{array}{ccc}1& 2& 3\\ 1/2& 1& 3\\ 1/3& 1/3& 1\end{array}\right];D_2=\left[\begin{array}{ccccc}1& 3& 1/4& 1/3& 3\\ 1/3& 1& 1/4& 1/4& 1/3\\ 4& 4& 1& 3& 3\\ 3& 4& 1/3& 1& 3\\ 1/3& 3& 1/3& 1/3& 1\end{array}\right];$

$D_3=\left[\begin{array}{cccc}1& 5& 2& 2\\ 1/5& 1& 1/3& 1/2\\ 1/2& 3& 1& 2\\ 1/2& 2& 1/2& 1\end{array}\right];D_4=\left[\begin{array}{cccc}1& 4& 2& 3\\ 1/4& 1& 2& 2\\ 1/2& 1/2& 1& 2\\ 1/3& 1/2& 1/2& 1\end{array}\right]$

Below with D ₁for example,

By judgment matrix $D_1=\left[\begin{array}{ccc}1& 2& 3\\ 1/2& 1& 3\\ 1/3& 1/3& 1\end{array}\right]$ Can n=3 be obtained,

First product is done to the value of the element of the every a line in judgment matrix D:

$M_i=d_{i1}\&times;d_{i2}\&times;...\&times;d_{in}=\left[\begin{array}{c}1\&times;2\&times;3\\ 1/2\&times;1\&times;3\\ 1/3\&times;1/3\&times;1\end{array}\right],$ N is judgment matrix exponent number, M _irepresent i-th key element importance ratio that is relative and other key elements long-pending, then to M _iopen n power, obtain

$W_i=\left[\begin{array}{c}\sqrt[3]{1\&times;2\&times;3}\\ \sqrt[3]{1/2\&times;1\&times;3}\\ \sqrt[3]{1/3\&times;1/3\&times;1}\end{array}\right]=\left[\begin{array}{c}1.817\\ 1.145\\ 0.4807\end{array}\right],$ Again to W _ibe normalized

score weight between note A1, A2, A3 tri-groups is w _a, namely $w_A=\left[\begin{array}{c}0.5278\\ 0.3325\\ 0.1396\end{array}\right];$

${\mathrm{\&lambda;}}_{max}=\underset{i=1}{\overset{n}{\&Sigma;}}\frac{{\left(Dw\right)}_i}{{\mathrm{nw}}_i}=3.0415;$

C _I＝(λ _max－n)/(n－1)＝(3.0415-3)/(3-1)＝0.0208；

When n=3 is, R _i=0.58,

C _r=0.0208/0.58=0.0358<0.1, then the consistance of D is acceptable.

Therefore, the weight of three first class index is respectively 0.5278, and 0.3325,0.1396.

In like manner, the weight in A1, A2, A3 tri-first-level class groups can be obtained, use w _a1, w _a2, w _a3represent, the weight of all two-level index equals the weight of each two-level index in group and the product of its place first class index shared weight between group.

Finally can determine that the weight of each index is as shown in the table.

Further, described step S2, it comprises:

S21, data envelope analysis specific algorithm refer to: wherein V=(v ₁, v ₂..., v _m) be the weight that DEA inputs, m is input variable quantity; U=(u ₁, u ₂..., u _k) be the weight that DEA exports, k is output variable quantity; Note $X_j=\left(\begin{array}{c}{x}_{1j}\\ x_{2j}\\ ...\\ x_{mj}\end{array}\right);Y_j=\left(\begin{array}{c}{y}_{1j}\\ y_{2j}\\ ...\\ y_{kj}\end{array}\right),$ J=1,2 ..., 13, because the storage system that greenness is higher wished by each grain depot, therefore, with " poor " (v ₁), " in " (v ₂) be the input of system, " good " (u ₁), " excellent " (u ₂) as the output of system, i.e. V=(v ₁, v ₂) _,u=(u ₁, u ₂).Wherein X _jfor the score matrix of system input, Y _jfor the score matrix that system exports.

$\left\{\begin{array}{cc}\frac{U\&CenterDot;Y}{V\&CenterDot;X}\&le;1,& (k=1,2,...,n)\\ V\&CenterDot;X=1& \\ U\&GreaterEqual;0& \\ V\&GreaterEqual;0& \end{array}\right.$

$\max X_{ij}=\frac{U\&CenterDot;Y}{V\&CenterDot;X}$

Wherein maxX _ijfor the performance of grain depot i on index j.

Select 10 experts to pass judgment on the first and second the third fourth four grain depots, assuming that the weight of each expert is equal, following table draw the basic data that DEA calculates, in following table, digital 0-8 refers to that 10 experts are to the votes of first to fourth grain depot 13 indexs four grades.

Due to the chemical technology that each grain depot has been wished, therefore, will with " poor " (v ₁), " in " (v ₂) be the input of system, " good " (u ₁), " excellent " (u ₂) as the output of system, for non-chemically Pesticidal technology, set up linear programming model, namely

$\left\{\begin{array}{c}0v_1+3v_2-6u_1-u_2\&GreaterEqual;0\\ 0v_1+2v_2-6u_1-2u_2\&GreaterEqual;0\\ 1v_1+4v_2-4u_1-1u_2\&GreaterEqual;0\\ 2v_1+3v_2-5u_1-0u_2\&GreaterEqual;0\\ v_1,v_2,u_1,u_2\&GreaterEqual;0\end{array}\right.$

For grain depot first, have

MaxX _{first 1}=6u ₁+ u ₂

Therefore $\left\{\begin{array}{c}0v_1+3v_2-6u_1-u_2\&GreaterEqual;0\\ 0v_1+2v_2-6u_1-2u_2\&GreaterEqual;0\\ 1v_1+4v_2-4u_1-1u_2\&GreaterEqual;0\\ 2v_1+3v_2-5u_1-0u_2\&GreaterEqual;0\\ 0v_1+3v_2=1\\ v_1,v_2,u_1,u_2\&GreaterEqual;0\end{array}\right.,$ Can be drawn by software lingo11 $\left\{\begin{array}{c}{v}_1=0\\ v_2=1/3\\ u_1=0.111111\\ u_2=0\end{array}\right.$

So, maxX _{first 1}=6 × 0.111111+0=0.6667

Wherein maxX ₁₁for the performance of grain depot first on non-chemically Pesticidal technology, maxX can be obtained _{first 1}=0.6667, in like manner can obtain the performance of other grain depots in other factors.

Further, described step S3, it comprises:

S31, carry out comprehensive evaluation in conjunction with index weights and refer to: the performance G of each grain depot in every evaluation index determined according to data envelope analysis _i={ maxX _i1, maxX _i2..., maxX _i13, wherein i refers to grain depot i, maxX _ijrefer to i-th performance of grain depot in a jth index, as shown in the table,

Two-level index	First	Second	Third	Fourth
					B1	0.6667	1.0000	0.3333	0.5556
B2	1.0000	0.6667	1.0000	1.0000
					B3	1.0000	0.1143	1.0000	1.0000
B4	0.6667	0.8333	0.8333	1.0000
					B5	0.4167	0.6667	0.6667	1.0000
B6	0.1667	1.0000	0.5714	1.0000
					B7	0.7500	0.5000	1.0000	0.3333
B8	1.0000	0.4444	0.3333	1.0000
					B9	0.6667	0.6667	0.5000	1.0000
B10	0.7500	0.6000	0.4800	1.0000
					B11	0.1786	0.2381	0.0857	1.0000
B12	0.5000	0.1875	0.0156	1.0000
					B13	1.0000	0.1143	1.0000	1.0000

Combine the weight sets L={w determined according to analytical hierarchy process again _b1, w _b2..., w _b13, w _bibe the weight of i-th factor, calculate the comprehensive evaluation score of each grain depot foodstuffs storage greenness, by : A _first=0.6667 × 0.0790+1.0000 × 0.0310+1.0000 × 0.2262+0.6667 × 0.1376+0.4167 × 0.0540+0.1667 × 0.1496+0.7500 × 0.0302+1.0000 × 0.0931+0.6667 × 0.0595+0.7500 × 0.0673+0.1786 × 0.0304+0.5000 × 0.0256+1.0000 × 0.0163=0.6895

In like manner can obtain: A _second=0.5762, A _third=0.6623, A _fourth=0.9446.

Therefore, the foodstuff preservation greenness comprehensive grading of 4 grain depots sorts as follows from low to high: second < third < first < fourth.Can also find that grain depot first is weaker in " grain depot layout " according to upper table in addition, and lack coming into operation of " recirculation fumigation equipment "; Grain depot second lacks the introduction of " advanced dry technology " and the Rational Maintenance to grain depot; Grain depot third needs to improve " auxiliary facility " and builds and strengthen " grain depot layout "; Grain depot fourth is badly in need of the improvement of " paddy drying equipment " aspect.

Claims (9)

1., based on a foodstuff preservation Green Degree Evaluation of DEA-AHP, it is characterized in that, comprise the following steps:

S1, according to View of Grain Storage Technoiogy requirement, set up foodstuff preservation greenness assessment index system with analytical hierarchy process (AHP); Use AHP method, organize the relative importance of expert's Comparative indices between two, provide corresponding proportional roles, build upper strata key element to the judgment matrix of lower floor's coherent element, and calculate eigenvalue of maximum and the character pair vector of judgment matrix, consistency check is passed through to the judgment matrix obtained;

Wherein, the index system of described foundation is 3 first class index and 13 two-level index;

Described development of judgment matrix concrete grammar is: according to hierarchy Model according to comparing scale development of judgment matrix respectively to element development of judgment matrix in index system group and between group, obtain 4 judgment matrixs altogether;

The eigenvalue of maximum of described judgment matrix is calculated by D × X=λ × X, and wherein λ is the eigenwert of matrix D, and non-vanishing vector X is called the proper vector corresponding to eigenvalue λ of D; The Maximum characteristic root λ of D existence anduniquess _max, adopt root method to calculate λ _maxwith the approximate value of W, then to W _ibe normalized, wherein i is i-th key element, namely wherein W is the weight vectors of each factor in D, w _ifor the weight vectors of each factor after normalization; Finally try to achieve the eigenvalue of maximum λ of judgment matrix _max, n is the exponent number of matrix D;

Described consistency check refers to: by formula C _r=C _i/ R _ithe consistance of D is tested, in formula: C _ifor consistency check index, C _i=(λ _max-n)/(n-1); N is the exponent number of judgment matrix; R _ifor Aver-age Random Consistency Index;

S2, determine the concrete manifestation of the corresponding each index of each foodstuff preservation scheme in conjunction with data envelope analysis (DEA);

If V=is (v ₁, v ₂..., v _m) be the input of DEA, m is input variable quantity; U=(u ₁, u ₂..., u _k) be the output of DEA, k is output variable quantity; Note j=1,2 ..., 13; Wherein X _jfor the score matrix of system input, Y _jfor the score matrix that system exports, that is to say the quantity of the evaluation result of expert, by following formula:

Finally obtain the performance of each grain depot in 13 indexs, wherein UY is u ₁y _1j+ u ₂y _2j+ ... u _ky _mj, VX is v ₁x _1j+ v ₂x _2j+ ... v _kx _mj, maxX _ijfor the performance of grain depot i on index j;

The index weights that S3, last binding hierarchy analytic approach and data envelope analysis draw carries out comprehensive evaluation to multiple foodstuff preservation scheme;

According to the weight G of each grain depot in every evaluation index that data envelope analysis is determined _i={ maxX _i1, maxX _i2..., maxX _i13, wherein i here refers to grain depot i, maxX _ijrefer to i-th performance of grain depot in a jth index, in conjunction with the weight sets L={w of 13 indexs determined according to analytical hierarchy process _b1, w _b2..., w _b13, w _bibe the weight of i-th factor, calculate the comprehensive evaluation score of each grain depot foodstuffs storage greenness, by comprehensive evaluation score can be calculated, wherein, A _ifor the comprehensive evaluation score of grain depot i foodstuff preservation greenness.

2. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, it is characterized in that, in step S1,3 described first class index are: green storage advanced technology A1, and green storage equipment and device A2 consider A3 with green storage.

3. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 2, it is characterized in that, wherein said green storage advanced technology A1 comprises: non-chemically Pesticidal technology B1, fungus control technology B2, advanced dry technology B3, air-conditioning storing grain technology B4 and low temperature treatment technology B5;

Described green storage equipment and device A2 comprise: recirculation fumigation equipment B6, paddy drying equipment B 7, mechanical ventilator B8 and grain condition monitoring device B9;

Described green storage considers A3 and comprises: rationally B13 safeguarded by grain depot addressing B10, rationally grain depot layout B11, Rational Matching facility B12 and reasonable grain depot.

4. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, it is characterized in that, the D of development of judgment matrix described in step S1 is specially, respectively to group described in and group between element development of judgment matrix, namely respectively with first class index A1, A2, A3 for criterion, compare between two between group, obtain a judgment matrix D ₁; Then in group, respectively with each index for criterion, compare between two, judge and the relation of other indexs, obtain judgment matrix D ₂, D ₃, D ₄, obtain 4 matrixes altogether; The comparison scale of described each index is:

(1) i with j is identical important, standard value d _ijbe 1;

(2) i is important in j a little, standard value d _ijbe 3;

(3) i is obviously important in j, standard value d _ijbe 5;

(4) i is strongly important in j, standard value d _ijbe 5;

(5) i is definitely important in j, standard value d _ijbe 9;

(6) i and j is in the intermediate value of (1)-(5) relation, standard value d _ijbe 2,4,6 or 8;

(7) the affinity criterions value d of a jth key element and i-th key element _jibe then 1/d _ij;

Wherein i and j represents i-th key element and a jth key element in table respectively, d _ijrefer to the affinity criterions value of i-th key element and a jth key element.

5. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, is characterized in that, adopts root method to calculate λ described in step S1 _maxbe specially with the approximate value of W:

First product M is done to the value of the element of the every a line in judgment matrix D _i=d _i1× d _i2× ... × d _in, n is judgment matrix exponent number, M _irepresent i-th key element importance ratio that is relative and other key elements long-pending, then to M _iopen n power, obtain

6. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, it is characterized in that, in the weight of each factor described in step S1, wherein the weight of all two-level index equals the weight of each two-level index in group and the product of its place first class index shared weight between group.

7. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, is characterized in that, the R of Aver-age Random Consistency Index described in step S1 _ivalue is judgment matrix exponent number when being 1 or 2, R _i=0; When judgment matrix exponent number is 3, R _i=0.58; When judgment matrix exponent number is 4, R _i=0.90;

When judgment matrix exponent number is 5, R _i=1.12; When judgment matrix exponent number is 6, R _i=1.24; When judgment matrix exponent number is 7, R _i=1.32; When judgment matrix exponent number is 8, R _i=1.41; When judgment matrix exponent number is 9, R _i=1.45.

8. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, is characterized in that, described in step S1 to the judgment matrix obtained by consistency check, wherein work as C _r<0.1, then the consistance of D accepts, and the w obtained is the weight vectors of each factor; Otherwise D need be readjusted to meeting consistency check requirement.

9. a kind of foodstuff preservation Green Degree Evaluation based on DEA-AHP according to claim 1, is characterized in that, DEA described in step S2 is input as " poor " (v ₁), " in " (v ₂); The output of DEA is " good " (u ₁), " excellent " (u ₂), i.e. V=(v ₁, v ₂), U=(u ₁, u ₂).