呂煒等
摘要:一些大型或技術(shù)復雜的政府工程確定贏標人時除投標價格因素外,還將質(zhì)量、工期等因素納入考慮。為此,文章將政府工程招投標時招標人、競標人的參與選擇過程視為一個要約、承諾和定標的三階段動態(tài)博弈,并構(gòu)建了政府工程多屬性招投標非合作動態(tài)博弈模型。然后將投標價格視為競標質(zhì)量和工期的函數(shù),求出了投標人的最優(yōu)競標戰(zhàn)略。通過分析發(fā)現(xiàn)投標質(zhì)量和提前工期分別超過某點后,投標人的收益分別是其投標質(zhì)量和提前工期的增函數(shù),而在到達該點之前,投標質(zhì)量和提前工期邊際收益遞減;招標人的投標承諾質(zhì)量和提前工期偏好系數(shù)越大,競標人的收益越高,其結(jié)果是招標人故意夸大其質(zhì)量、工期偏好系數(shù),以誘導競標人投標,增大招標剩余。最后通過算例驗證了模型的有效性。
關(guān)鍵詞:政府工程;非合作博弈;多屬性;招投標;博弈
中圖分類號:F28 文獻標志碼:A 文章編號:
10085831(2014)02006409
招投標最早起源于18世紀的英國,它通過制定一套明確、具體的交易制度,并通過招投標參與人的報價來確定中標人和成交價格,以合理配置資源。中國有史料記載的招投標發(fā)生在1902年張之洞創(chuàng)辦的湖北皮革廠,但解放前封建、半封建和半殖民地社會制度束縛了招投標事業(yè)的發(fā)展。新中國成立后,中國又實行計劃經(jīng)濟,招投標一度停止。20世紀80年代以來,中國逐步推行招投標制度,并首先在工程建設領(lǐng)域予以推行,取得了顯著成效。中國加入世界貿(mào)易組織后,為了與國際接軌,于2003年開始推行工程量清單計價招標,并采取國外的最低價格中標制度。但近年來,隨著中國經(jīng)濟的快速發(fā)展,政府財力日益增強,政府投資大型基礎設施項目和公益項目日益增加,而這些大型項目建設規(guī)模較大、技術(shù)復雜、建設周期長、資金占用量大,不確定影響因素較多,所需設備品種多、價值量高,因此其質(zhì)量、工期和成本直接關(guān)系到整個項目的成敗[1]。在這類政府工程招投標中,招標人不僅關(guān)注投標報價,還關(guān)注投標人的工程質(zhì)量、工期等信息(包括競標人資金、信用、類似工程業(yè)績等)。傳統(tǒng)的招投標(逆向拍賣)僅將競標人的投標報價作為定標依據(jù),而將工程投標質(zhì)量和工期的信息作為資格審查條件,可能導致部分性價比較高的競標人(指質(zhì)量高、工期短且報價相對較低的競標人)失去中標機會,既不利于節(jié)約政府工程采購費用,也不利于質(zhì)優(yōu)價廉競標人體現(xiàn)其競爭優(yōu)勢,難以滿足一些大型或技術(shù)復雜的政府工程招標的需要,因此有必要采用多屬性招標。
多屬性拍賣是傳統(tǒng)拍賣的拓展,Bichler認為多屬性拍賣是招標人與競標人進行交易時考慮拍賣品的多個品質(zhì)屬性(如質(zhì)量、交貨期等)的一種拍賣方式[2]。Thiel最早研究多屬性拍賣,研究如何將多屬性拍賣轉(zhuǎn)化為傳統(tǒng)拍賣問題,發(fā)現(xiàn)在已知賣方偏好函數(shù)的情況下,多拍拍賣問題可以簡化為一維采購拍賣問題[3]。根據(jù)拍賣行動規(guī)則不同,多屬性拍賣又分為菜單式拍賣、選美式拍賣和記分拍賣[4]。Asker & Cantillon證明記分制拍賣占優(yōu)于菜單式和選美式拍賣[5]。Che考慮了包括價格和質(zhì)量的二維拍賣模型,針對贏標人的確定和支付問題,研究了密封拍賣機制的變形協(xié)議,即第一記分拍賣、第二記分拍賣和第二首選要約記分拍賣[6]。Branco將Che的研究拓展到競標人成本相互關(guān)聯(lián)的情形,發(fā)現(xiàn)成本的相關(guān)性對多維拍賣最優(yōu)機制有重要影響[7]。David研究了具有普遍意義的多屬性拍賣[8]。王宏通過引入廣義質(zhì)量生產(chǎn)函數(shù),求解了多維信息招投標的最優(yōu)招標機制[1]。黃河假設競標人對質(zhì)量和價格分別投標,并利用樹形結(jié)構(gòu)求解競勝標問題[9]。孫亞輝在David研究的基礎上研究多屬性拍賣方式下競標人的最優(yōu)投標策略[10]。李軍和劉樹林研究了基于C-D效用函數(shù)的多屬性采購拍賣[11]。周學廣運用博弈理論研究了供應鏈在線多屬性逆向拍賣,但其僅考慮招標人效用函數(shù)視為質(zhì)量、交貨期的線性函數(shù)的情形[12]。
一、模型
(一)問題描述
不妨考慮政府工程建設業(yè)主針對某單位工程采取多屬性招標確定建筑承包商。招標文件規(guī)定投標人需要投標的多屬性包括:報價、質(zhì)量等級和工期。政府工程招投標流程有資格預審文件、招標文件的編制與審查、發(fā)布招標公告(或投標邀請書)、資格預審、出售招標文件、勘察現(xiàn)場、投標預備會議、投標文件的編制與遞交、工程標底價格的報審、開標、評標、中標和簽訂合同。從博弈視角可將政府工程建設業(yè)主和承包商的招投標過程視為一個三階段的動態(tài)博弈,其博弈流程如圖1。第一階段,建設業(yè)主發(fā)布某建設單位(項)工程招標公告,以及該項目建設地址、規(guī)模、詳細的需求標準(招標圖紙等)、工程開工日期,并提供招標工程量清單細目明細,以利于投標人評估測算是否參與該工程競標。第二階段,建筑承包商根據(jù)業(yè)主提供的招標文件、相關(guān)建設標準、工程量清單計價定額和設備材料市場價格等數(shù)據(jù)信息,并結(jié)合自身的施工成本、質(zhì)量以及施工工期,決定其是否參與投標以及投標競爭策略。第三階段,定標結(jié)束,建設業(yè)主或委托招標人組織評標委員會按照招標文件要求,對投標人的投標文件進行篩選和評估,以選擇最佳的投標文件作為決標和授予施工合同(為簡化分析,此處采取與國際招標通行的低價中標,即報價最低的投標人能夠給建設業(yè)主帶來最大的利潤),招標結(jié)束。因此可以將政府工程招投標視為一個多階段的擴展式博弈,如果將所有投標人視為一個參與人,則在政府工程招標的每個階段只有一個參與人采取行動,且其具有非簡單選擇集——選擇集里的元素個數(shù)大于1,而其他參與人則僅有一個單元素,即“不采取任何行動”的選擇集,所以可以將政府工程招投標視為具有完美信息的多階段博弈。
圖1 政府工程招投標博弈
(二)模型基本假設
四、結(jié)論
本文將政府工程多屬性招投標視為一個招標要約、投標承諾和定標的三階段動態(tài)博弈問題,并構(gòu)建了非合作動態(tài)博弈模型。該模型考慮了投標價格、工期和質(zhì)量三個屬性,將競標人的類型函數(shù)通過質(zhì)量和工期予以表示,并將投標價格視為質(zhì)量、工期的函數(shù),利用逆推法求解了該模型子博弈納什均衡。在子博弈納什均衡時,競標人會根據(jù)自身真實的施工質(zhì)量和工期進行投標,并進一步得出招標人剩余效用和競標人收益分別與投標質(zhì)量、工期的圖形呈U形,即投標質(zhì)量和提前工期分別超過某點后,投標人的收益分別是其投標質(zhì)量和提前工期的增函數(shù);而在到達該點之前,投標質(zhì)量和提前工期邊際收益遞減;招標人的投標承諾質(zhì)量和提前工期偏好系數(shù)越大,競標人的收益越高,因此競標人的投標也越積極,招標人的剩余也越大;競標人施工質(zhì)量和工期成本系數(shù)值越大,招標人剩余越小。
論文構(gòu)建的政府工程多屬性招投標博弈模型是建立在公開招標人偏好系數(shù)的基礎上,而現(xiàn)實政府工程招投標信息通常難以滿足此條件,因此針對此問題可進一步深入研究。參考文獻:
[1]王宏, 陳宏民, 楊劍俠. 多維信息招投標中的最優(yōu)機制及其實施[J].管理科學學報,2010,13(08):1-14.
[2]BICHLER M. An experimental analysis of multiattribute auctions[J].Decision Support Systems,2000,29(3): 249-268.
[3]THIEL S E. Multidimensional auctions [J]. Economics Letters,1988,28(1):37-40.
[4]唐邵玲, 劉琳. 兩種記分函數(shù)下多屬性拍賣投標均衡策略[J].經(jīng)濟數(shù)學, 2011(2):54-59.
[5]ASKER J, CANTILLON E. Properties of scoring auctions [J]. The RAND Journal of Economics,2008,39(1):69-85.
[6]CHE Y K. Design competition through multidimensional auctions[J]. rand Journal of Economics,1993, 24(6):68-80.
[7]BRANCO F. The design of multidimensional auctions [J]. The RAND Journal of Economics,1997,20:63-81.
[8]DAVID E, AZOULAYSCHWARTZ R, KRAUS S. Bidding in sealedbid and English multiattribute auctions[J]. Decision Support Systems,2006,42(2):527-56.
[9]黃河, 徐鴻雁, 陳劍. 多因素采購組合拍賣獲勝者確定問題研究[J].系統(tǒng)工程理論與實踐,2008,28(7):27-33.
[10]孫亞輝, 馮玉強. 多屬性密封拍賣模型及最優(yōu)投標策略[J].系統(tǒng)工程理論與實踐,2010(7):1185-9.
[11]李軍, 劉樹林. 基于Cobb-Douglas效用函數(shù)的多屬性采購拍賣[J].管理科學學報,2012(3):54-60.
[12]周學廣, 張堅, 梅強, 等. 基于多屬性逆向拍賣的博弈分析[J].管理工程學報,2011,25(2):200-5.
[13]FUDENBERG D, TIROLE J. Game theory. 1991 [M]. Cambridge,MA:MIT Press,1991.
A Game Analysis Based on
Multiattribute Bidding of Government Project
LYU Wei, HE Changzheng
(School of Business, Sichuan University, Chengdu 610074, P. R. China)
Abstract:
Determining the winning bidder of some major or technically complex government project is not only the price, bat also the quality and duration etc. So, the paper regards the selection process of government project tenderee and bidder as a threestage dynamic game of offer, commitment and evaluation, and to build a noncooperative and dynamic game model based on multiattribute government project bidding, and then to consider the price of a function of quality and duration, to find out the optimal bidding strategy of bidder. The analysis turns up that the quality and advanced time respectively after reach one point, the bidders earnings are a rise function of quality and advance time, and then the bidders earnings would be a decrease function if it before reaches the point. The bidders earnings would be much more if the preferences coefficient is much greater of quality and advance time by which tenderee promises, so to gain more bidding surplus, the tenderee would exaggerate the preferences coefficient of quality and advance times. At last, this paper verifies the validity of the model by using examples.
Key words: government project; noncooperative game; multiattribute; bidding; game
(責任編輯 傅旭東)
論文構(gòu)建的政府工程多屬性招投標博弈模型是建立在公開招標人偏好系數(shù)的基礎上,而現(xiàn)實政府工程招投標信息通常難以滿足此條件,因此針對此問題可進一步深入研究。參考文獻:
[1]王宏, 陳宏民, 楊劍俠. 多維信息招投標中的最優(yōu)機制及其實施[J].管理科學學報,2010,13(08):1-14.
[2]BICHLER M. An experimental analysis of multiattribute auctions[J].Decision Support Systems,2000,29(3): 249-268.
[3]THIEL S E. Multidimensional auctions [J]. Economics Letters,1988,28(1):37-40.
[4]唐邵玲, 劉琳. 兩種記分函數(shù)下多屬性拍賣投標均衡策略[J].經(jīng)濟數(shù)學, 2011(2):54-59.
[5]ASKER J, CANTILLON E. Properties of scoring auctions [J]. The RAND Journal of Economics,2008,39(1):69-85.
[6]CHE Y K. Design competition through multidimensional auctions[J]. rand Journal of Economics,1993, 24(6):68-80.
[7]BRANCO F. The design of multidimensional auctions [J]. The RAND Journal of Economics,1997,20:63-81.
[8]DAVID E, AZOULAYSCHWARTZ R, KRAUS S. Bidding in sealedbid and English multiattribute auctions[J]. Decision Support Systems,2006,42(2):527-56.
[9]黃河, 徐鴻雁, 陳劍. 多因素采購組合拍賣獲勝者確定問題研究[J].系統(tǒng)工程理論與實踐,2008,28(7):27-33.
[10]孫亞輝, 馮玉強. 多屬性密封拍賣模型及最優(yōu)投標策略[J].系統(tǒng)工程理論與實踐,2010(7):1185-9.
[11]李軍, 劉樹林. 基于Cobb-Douglas效用函數(shù)的多屬性采購拍賣[J].管理科學學報,2012(3):54-60.
[12]周學廣, 張堅, 梅強, 等. 基于多屬性逆向拍賣的博弈分析[J].管理工程學報,2011,25(2):200-5.
[13]FUDENBERG D, TIROLE J. Game theory. 1991 [M]. Cambridge,MA:MIT Press,1991.
A Game Analysis Based on
Multiattribute Bidding of Government Project
LYU Wei, HE Changzheng
(School of Business, Sichuan University, Chengdu 610074, P. R. China)
Abstract:
Determining the winning bidder of some major or technically complex government project is not only the price, bat also the quality and duration etc. So, the paper regards the selection process of government project tenderee and bidder as a threestage dynamic game of offer, commitment and evaluation, and to build a noncooperative and dynamic game model based on multiattribute government project bidding, and then to consider the price of a function of quality and duration, to find out the optimal bidding strategy of bidder. The analysis turns up that the quality and advanced time respectively after reach one point, the bidders earnings are a rise function of quality and advance time, and then the bidders earnings would be a decrease function if it before reaches the point. The bidders earnings would be much more if the preferences coefficient is much greater of quality and advance time by which tenderee promises, so to gain more bidding surplus, the tenderee would exaggerate the preferences coefficient of quality and advance times. At last, this paper verifies the validity of the model by using examples.
Key words: government project; noncooperative game; multiattribute; bidding; game
(責任編輯 傅旭東)
論文構(gòu)建的政府工程多屬性招投標博弈模型是建立在公開招標人偏好系數(shù)的基礎上,而現(xiàn)實政府工程招投標信息通常難以滿足此條件,因此針對此問題可進一步深入研究。參考文獻:
[1]王宏, 陳宏民, 楊劍俠. 多維信息招投標中的最優(yōu)機制及其實施[J].管理科學學報,2010,13(08):1-14.
[2]BICHLER M. An experimental analysis of multiattribute auctions[J].Decision Support Systems,2000,29(3): 249-268.
[3]THIEL S E. Multidimensional auctions [J]. Economics Letters,1988,28(1):37-40.
[4]唐邵玲, 劉琳. 兩種記分函數(shù)下多屬性拍賣投標均衡策略[J].經(jīng)濟數(shù)學, 2011(2):54-59.
[5]ASKER J, CANTILLON E. Properties of scoring auctions [J]. The RAND Journal of Economics,2008,39(1):69-85.
[6]CHE Y K. Design competition through multidimensional auctions[J]. rand Journal of Economics,1993, 24(6):68-80.
[7]BRANCO F. The design of multidimensional auctions [J]. The RAND Journal of Economics,1997,20:63-81.
[8]DAVID E, AZOULAYSCHWARTZ R, KRAUS S. Bidding in sealedbid and English multiattribute auctions[J]. Decision Support Systems,2006,42(2):527-56.
[9]黃河, 徐鴻雁, 陳劍. 多因素采購組合拍賣獲勝者確定問題研究[J].系統(tǒng)工程理論與實踐,2008,28(7):27-33.
[10]孫亞輝, 馮玉強. 多屬性密封拍賣模型及最優(yōu)投標策略[J].系統(tǒng)工程理論與實踐,2010(7):1185-9.
[11]李軍, 劉樹林. 基于Cobb-Douglas效用函數(shù)的多屬性采購拍賣[J].管理科學學報,2012(3):54-60.
[12]周學廣, 張堅, 梅強, 等. 基于多屬性逆向拍賣的博弈分析[J].管理工程學報,2011,25(2):200-5.
[13]FUDENBERG D, TIROLE J. Game theory. 1991 [M]. Cambridge,MA:MIT Press,1991.
A Game Analysis Based on
Multiattribute Bidding of Government Project
LYU Wei, HE Changzheng
(School of Business, Sichuan University, Chengdu 610074, P. R. China)
Abstract:
Determining the winning bidder of some major or technically complex government project is not only the price, bat also the quality and duration etc. So, the paper regards the selection process of government project tenderee and bidder as a threestage dynamic game of offer, commitment and evaluation, and to build a noncooperative and dynamic game model based on multiattribute government project bidding, and then to consider the price of a function of quality and duration, to find out the optimal bidding strategy of bidder. The analysis turns up that the quality and advanced time respectively after reach one point, the bidders earnings are a rise function of quality and advance time, and then the bidders earnings would be a decrease function if it before reaches the point. The bidders earnings would be much more if the preferences coefficient is much greater of quality and advance time by which tenderee promises, so to gain more bidding surplus, the tenderee would exaggerate the preferences coefficient of quality and advance times. At last, this paper verifies the validity of the model by using examples.
Key words: government project; noncooperative game; multiattribute; bidding; game
(責任編輯 傅旭東)