基于模糊綜合評(píng)判法的新產(chǎn)品開發(fā)方案評(píng)價(jià)

基于模糊綜合評(píng)判法的新產(chǎn)品開發(fā)方案評(píng)價(jià)

隨著科學(xué)技術(shù)的快速發(fā)展，新產(chǎn)品的開發(fā)具有特殊的重要性。運(yùn)用模糊數(shù)學(xué)的一種具體應(yīng)用方法——模糊綜合評(píng)判法,結(jié)合可信因子綜合方法,可提高新產(chǎn)品開發(fā)的成功率。1模型1:可信因子模糊性是事物相對(duì)性的一種數(shù)學(xué)描述,模糊集是具有某種性質(zhì)的界限不明確的事物和現(xiàn)象的集合。隸屬度是指一個(gè)元素屬于某一模糊集合的程度或資格。模糊矩陣又稱模糊關(guān)系矩陣。模糊關(guān)系是指事實(shí)上存在,但又不能清晰確定,不能用普通有序?qū)厦枋龅年P(guān)系。設(shè)U=(u1,u2,…,un),V=(v1,v2,…,vm),rij代表元素集合描述ui和vj具有模糊關(guān)系的程度,則從U到V的模糊關(guān)系矩陣為:R=[r11r12?r1mr21r22?r2m???rn1rn2?rnm]這里:U為評(píng)價(jià)因素集;V為抉擇語等級(jí)集;rij為單獨(dú)考慮因素ui時(shí),ui的評(píng)價(jià)對(duì)等級(jí)vj的錄屬度;R又稱為總評(píng)價(jià)矩陣。模糊變換B=A?R=(b1?b2???bm)為綜合評(píng)價(jià)的數(shù)學(xué)模型。式中:A是各評(píng)判因素權(quán)重的向量;R為各評(píng)判因素對(duì)各評(píng)語的隸屬度;B是綜合評(píng)判的結(jié)果。R中元素rij既代表ui關(guān)于vj的“隸屬程度”,也反映了評(píng)判ui為vj的集中程度,可將其量化作為一個(gè)重要參數(shù)體現(xiàn)在最終的評(píng)判結(jié)果中。稱反映這種集中程度的量為“可信因子”或“信度”。權(quán)數(shù)A的確定方法可采用Dilphy法、FD法、判斷矩陣法、AHP法等。確定信度的方法可用表1表示。其中等級(jí)1可代表十分重要,…,等級(jí)N為極不重要等。等級(jí)量化后,對(duì)應(yīng)于閉區(qū)間上N次平分。如N取5,則依次得到[0.0,0.2],[0.2,0.4],[0.4,0.6],[0.6,0.8],[0.8,1.0]。要求每個(gè)評(píng)判者對(duì)ui(i=1,2,…,n)的重要性等級(jí)給出選擇,填入重要性等級(jí)表相應(yīng)欄中,對(duì)各欄票數(shù)進(jìn)行統(tǒng)計(jì),再進(jìn)行歸一化處理。例如u3代表產(chǎn)品競爭能力,統(tǒng)計(jì)出各等級(jí)票數(shù)pij(j=1,2,…,n)為(3,6,0,0,0),歸一化得數(shù)組{pi1/k,pi2/k,…,pin/k},即(3/9,6/9,0,0,0)?？扇?quán)重向量ai的信度max(pij/k)(j=1,2,…,n)。對(duì)權(quán)重向量A={a1,a2,?,an}(n∑i=1ai=1)就有對(duì)應(yīng)的信度向量{c1,c2,…,cn}。下面對(duì)可信因子進(jìn)行綜合。對(duì)于B=A。R={b1,b2,…,bm},其信度合成為:Ηi=εmax{θ1i,θ2i,?,θmi}+[(1-ε)/n]∑θjii=1,2,?,m(1)其中:θij=εmin{cj,rij}+(1-ε)(cj+rij)/2j=1,2,??n(2)ε可如下配置:ε=1-min{c1,c2,…,cn}得到信度的評(píng)判結(jié)果:ˉB={(ˉb1,Η1),(ˉb2,Η2),?,(ˉbm,Ηm)}當(dāng)出現(xiàn)多峰值時(shí),按最大隸屬度法難以確定,但通過可信因子綜合,可判定歸屬等級(jí)。在新產(chǎn)品開發(fā)決策中,利用綜合評(píng)判模型進(jìn)行綜合評(píng)判,可避免企業(yè)只考慮利潤最大化,而是將綜合效益作為企業(yè)目標(biāo),并且使決策者一開始就從整體角度對(duì)該種新產(chǎn)品開發(fā)進(jìn)程有較深刻的了解,對(duì)整個(gè)產(chǎn)品經(jīng)濟(jì)壽命周期中可能出現(xiàn)的風(fēng)險(xiǎn)及早防范。為了使評(píng)價(jià)結(jié)果更加可信,在一般模糊評(píng)判模型的基礎(chǔ)上引進(jìn)“可信因子”及其綜合方法,體現(xiàn)了模糊綜合評(píng)判不確定性的實(shí)質(zhì)。帶信度的評(píng)判結(jié)果使一些原本不易判定的問題得到合理解決,同時(shí)也提醒決策者慎重對(duì)待信度低的評(píng)判結(jié)果。2要素評(píng)判矩陣確定對(duì)黑龍江省某防火門廠開發(fā)的高層建筑防火門進(jìn)行開發(fā)方案的模糊綜合評(píng)價(jià)。1)設(shè)定開發(fā)防火門的評(píng)價(jià)要素。根據(jù)已知的影響新型防火門開發(fā)環(huán)境的各項(xiàng)指標(biāo)因素將評(píng)判要素分為3個(gè)層次,如圖1所示。2)用戶評(píng)判。評(píng)判等級(jí)定為5個(gè),即:評(píng)判等級(jí)={很好,好,一般,差,很差}。將5個(gè)產(chǎn)品評(píng)判要素列在一張表上,附產(chǎn)品功能介紹,分發(fā)至有關(guān)用戶進(jìn)行評(píng)判,采用專訪方式共發(fā)出評(píng)判表50份,得到有效評(píng)判表48份。對(duì)數(shù)據(jù)進(jìn)行分析整理后,得到各要素的評(píng)判結(jié)果如下:R質(zhì)=(0.3,0.4,0.3,0.0,0.0)R價(jià)=(0.1,0.4,0.3,0.1,0.1)R外=(0.2,0.5,0.2,0.1,0.0)R體=(0.1,0.4,0.5,0.0,0.0)R功=(0.2,0.3,0.5,0.0,0.0)請(qǐng)10位專家及本企業(yè)10位各部門負(fù)責(zé)人對(duì)其他評(píng)判要素進(jìn)行評(píng)判,共得到20份評(píng)判表。對(duì)數(shù)據(jù)分析整理后,得各要素的評(píng)判結(jié)果如下:R開發(fā)能力=(0.1,0.3,0.4,0.2,0.0)R生產(chǎn)技術(shù)=(0.0,0.2,0.5,0.2,0.1)R生產(chǎn)設(shè)備=(0.3,0.3,0.4,0.0,0.0)R組織管理=(0.1,0.6,0.3,0.0,0.0)R能=(0.2,0.3,0.3,0.1,0.1)R運(yùn)=(0.3,0.5,0.2,0.0,0.0)R承=(0.0,0.5,0.2,0.2,0.1)R投=(0.1,0.4,0.3,0.2,0.0)R獲=(0.4,0.3,0.3,0.0,0.0)R資=(0.4,0.5,0.1,0.0,0.0)R營=(0.0,0.3,0.3,0.1,0.1)R產(chǎn)=(0.4,0.3,0.3,0.0,0.0)R潛=(0.4,0.5,0.1,0.0,0.0)R環(huán)=(0.6,0.3,0.1,0.0,0.0)R社=(0.2,0.3,0.4,0.1,0.0)R連=(0.3,0.5,0.1,0.1,0.0)3)由10位專家和10位企業(yè)主要負(fù)責(zé)人采用德爾斐方法結(jié)合FD法確定各層次的權(quán)重向量分別為:A產(chǎn)=(0.40,0.30,0.05,0.05,0.20)A企=(0.2,0.2,0.1,0.3,0.1,0.1)A財(cái)=(0.3,0.2,0.5)A市=(0.2,0.3,0.3,0.2)A社=(0.4,0.3,0.3)A技=(0.6,0.4)A評(píng)=(0.30,0.40,0.25,0.05)4)模糊綜合評(píng)判。按照前述步驟,由質(zhì)量、價(jià)格、外觀、規(guī)格重量、功能適用性構(gòu)成要素評(píng)判矩陣:R產(chǎn)=[0.30.40.30.00.00.10.40.30.10.10.20.50.20.10.00.10.40.50.00.00.20.30.50.00.0]采用M(·,⊕)模型進(jìn)行綜合評(píng)判,得:B產(chǎn)=A產(chǎn)?R產(chǎn)=(0.205?0.385?0.345?0.035?0.030)類似的有B企=A企。R企=(0.13,0.39,0.36,0.09,0.03)。進(jìn)一步合成可得到技術(shù)性能指標(biāo)要素的評(píng)判矩陣:R技=[0.0250.3850.3450.0350.0300.1300.3900.3600.0900.030]采用M(·,⊕)模型進(jìn)行綜合評(píng)判,得:B技=A技?R技=(0.175?0.387?0.351?0.057?0.030)同理可得:B財(cái)=A財(cái)?R財(cái)=(0.22?0.38?0.27?0.10?0.03)B市=A市?R市=(0.28?0.38?0.28?0.03?0.03)B社=A社?R社=(0.39?0.36?0.19?0.06?0.06?0.00)由上可得到由技術(shù)性能要素、財(cái)務(wù)指標(biāo)要素、市場環(huán)境評(píng)價(jià)要素、社會(huì)生態(tài)評(píng)價(jià)要素組成的要素評(píng)判矩陣為:R評(píng)=[0.1750.3870.3510.0570.0300.2200.3800.2700.1000.0300.2800.3800.2800.0300.0300.3900.3600.1900.0600.000]采用M(∧,∨)模型進(jìn)行綜合評(píng)判,得到最終的新產(chǎn)品開發(fā)決策的評(píng)判結(jié)果:B評(píng)=A評(píng)?R評(píng)=(0.2300?0.3811?0.2928?0.0676?0.0285)5)可信分析。發(fā)放表1對(duì)技術(shù)性能要素、財(cái)務(wù)指標(biāo)要素、市場環(huán)境要素、社會(huì)生態(tài)評(píng)價(jià)要素確定信度如下:C技=12/20C財(cái)=14/20C市=7/20C社=6/20同時(shí)進(jìn)行信度的綜合,根據(jù)式(2)得:θij=[0.238750.419000.388000.138000.116000.292000.428000.335000.190000.130000.290500.355000.290000.078000.078000.313500.309000.207000.096000.04500]根據(jù)信度公式(1)得:H1=0.3046,H2=0.4129,H3=0.3633,H4=0.1707,H5=0.1190按隸屬度原則擬定方案列進(jìn)第二等級(jí)“好”類。根據(jù)可信因子的綜合