外文翻譯--夾具裝置的多目標(biāo)優(yōu)化設(shè)計(jì)

1、<p>　　外文資料名稱(chēng)：Multi-objective optimal fixture layout design </p><p>　　外文資料出處：Robotics and Computer Integrated Manufacturing </p><p>　　附 件： 1。外文資料翻譯全文 </p><p&

2、gt;　　2。外文原文 </p><p>　　夾具裝置的多目標(biāo)優(yōu)化設(shè)計(jì)</p><p>　　Diana M. Pelinescu, Michael Yu Wang*</p><p><b>　　陳少衛(wèi)譯</b></p><p>　　摘要: 本文重點(diǎn)闡述了夾具布局設(shè)計(jì)，主要用于多種質(zhì)量標(biāo)準(zhǔn)

3、下確定和評(píng)價(jià)夾具設(shè)計(jì)的可用性，并利用近似交換法結(jié)合多種性能要求進(jìn)行夾具優(yōu)化設(shè)計(jì)。重要的性能參數(shù)與運(yùn)動(dòng)定位的基本要求和封閉形式下的總體定位。三個(gè)性能參數(shù)為工件定位精確度、基準(zhǔn)和定位接觸力的分布。本文著重利用分級(jí)遞進(jìn)的方法進(jìn)行多基準(zhǔn)優(yōu)化設(shè)計(jì)。引入多種有效的算法并推廣應(yīng)用到各種實(shí)際案例，形成了執(zhí)行夾具組合的適當(dāng)交換方法。舉例說(shuō)明了上述方法的實(shí)驗(yàn)觀察和實(shí)驗(yàn)效果。</p><p>　　關(guān)鍵詞：夾具；夾具布局；夾具組合；多目

4、標(biāo)優(yōu)化設(shè)計(jì)</p><p><b>　　1. 引言</b></p><p>　　在產(chǎn)品加工精度和裝配精度方面，適當(dāng)?shù)膴A具設(shè)計(jì)是保證產(chǎn)品質(zhì)量的關(guān)鍵。夾具系統(tǒng)通常由夾緊和定位組成，必須能夠通過(guò)及其操作執(zhí)行定位并夾緊工件。雖然出現(xiàn)了一些像3-2-1規(guī)則設(shè)計(jì)指南，但基于CAD模型的自動(dòng)夾具設(shè)計(jì)系統(tǒng)一直發(fā)展緩慢。</p><p>　　本文論述了一類(lèi)三維工

5、件夾具的自動(dòng)設(shè)計(jì)方法。被夾緊工件是一個(gè)任意復(fù)雜的幾何體,夾具有最小數(shù)量的限位元素，即六個(gè)定位和一個(gè)夾緊。而且, 夾具要素被看作是非摩擦接觸點(diǎn)接觸，并限定于在具體的特定的定位。一般來(lái)說(shuō),一套有效的夾具可假設(shè)為一個(gè)大的收集裝置。例如,定位點(diǎn)可能產(chǎn)生于離散工件的表面。夾具設(shè)計(jì)的要求首先是確定可行夾具結(jié)構(gòu)，滿足定位和閉合的要求。其次, 該夾具設(shè)計(jì)需基于一些評(píng)價(jià)標(biāo)準(zhǔn)和最優(yōu)(或次優(yōu))夾具的選擇。這表現(xiàn)為定位精度、規(guī)范和分散定位力。這些要求包括了夾具

6、設(shè)計(jì)的關(guān)鍵技術(shù)。這些多重要求可能會(huì)沖突。因此，必須進(jìn)行多目標(biāo)夾具設(shè)計(jì)。</p><p>　　文中給出的是一種優(yōu)化夾具的設(shè)計(jì)方法。該方法是基于理論的最佳實(shí)驗(yàn)設(shè)計(jì)。 研制了一種高效評(píng)價(jià)受力遞歸特性的方法，自己設(shè)計(jì)開(kāi)發(fā)了受力分析軟件。優(yōu)化算法夾具設(shè)計(jì)制作了一套適應(yīng)多種性能要求的算法。</p><p><b>　　2.相關(guān)工作</b></p><p>

7、　　參考文獻(xiàn)[1]是普通夾具技術(shù)。多年來(lái)對(duì)夾具有了一個(gè)基本要求,近年來(lái)已廣泛應(yīng)用于自動(dòng)化領(lǐng)域。有幾種基于螺旋理論來(lái)分析夾具性能，來(lái)解決諸如封閉運(yùn)動(dòng)學(xué)，接觸類(lèi)型和摩擦效應(yīng)。</p><p>　　參考文獻(xiàn)[7]是基于不同的幾何分析技術(shù)。除了封閉運(yùn)動(dòng)學(xué)，組合夾具自動(dòng)設(shè)計(jì)程序是基于這一方法開(kāi)發(fā)的。近來(lái)固定模塊化設(shè)計(jì)問(wèn)題得到更多的關(guān)注。有深入研究夾具設(shè)計(jì)、圍繞工件和堅(jiān)固的夾具結(jié)構(gòu)，可行的刀具和路徑間隙。</p>

8、<p>　　綜合研究夾具問(wèn)題，已大致集中在夾具組成上，尤其在自動(dòng)化方面的應(yīng)用。參考文獻(xiàn)[14]描述的是夾具設(shè)計(jì)技術(shù)。</p><p><b>　　3. 夾具模型</b></p><p>　　夾具的基本性能在于被夾緊的運(yùn)動(dòng)約束。其運(yùn)動(dòng)條件是清楚的。如一個(gè)夾具有n個(gè)定位點(diǎn)，(i=1,2,…n),夾具可表示為：</p><p>　　由和

9、確定了定位位置和工件定位。夾具設(shè)計(jì)由定位矩陣來(lái)定義</p><p>　　，當(dāng)和，表面標(biāo)準(zhǔn)和工件表面接觸點(diǎn)的位置。夾具設(shè)計(jì)需要一個(gè)綜合計(jì)劃，以滿足一定的裝夾性能要求。</p><p>　　4. 夾具的質(zhì)量性能規(guī)范</p><p><b>　　4.1 精確定位</b></p><p>　　夾具質(zhì)量的一個(gè)重要方面就是工件定位在

10、夾具上的精度要高。一般工件的位置誤差由幾何變形和定位誤差造成的。本文只討論定位誤差造成的誤差。作為一種夾具模型的擴(kuò)展(公式(1))，定位誤差和工件定位誤差可表示如下：</p><p><b>　?。?）</b></p><p>　　顯然, 誤差來(lái)源于定位基準(zhǔn)的工件定位精度。定位誤差可表示為系統(tǒng)矩陣或是信息矩陣，M=GGT。實(shí)驗(yàn)表明，一個(gè)確定的標(biāo)準(zhǔn)達(dá)到高定位精度是信息矩

11、陣的最大決定因素(D-optimality)等等，最大值（即M）。</p><p><b>　　4.2最小定位力</b></p><p>　　夾具設(shè)計(jì)的另一個(gè)要求是盡量減少定位接觸力和任何約束或閉合力。讓我們討論一下完全確定的位置矢量和單元表面。對(duì)于點(diǎn)接觸，廣義的工件夾緊力可表示為：</p><p><b>　?。?）</b&g

12、t;</p><p>　　表示夾緊力的大小。因此,定位接觸力可表示為</p><p><b>　　（4）</b></p><p>　　我們可得出這些力的夾緊強(qiáng)度：</p><p>　　當(dāng) （5）</p><p>　　在閉合情況下,需要將這些力定位于N個(gè)定位點(diǎn)中的每個(gè)定

13、位點(diǎn)：</p><p>　　計(jì)算平均定位接觸力為：</p><p><b>　　（6）</b></p><p>　　得出一個(gè)適當(dāng)?shù)脑O(shè)計(jì)值（‖Pc‖）。這表明，在優(yōu)化設(shè)計(jì)過(guò)程中，應(yīng)該明確定位點(diǎn)和夾緊定位。</p><p>　　4.3平衡定位接觸力</p><p>　　夾具設(shè)計(jì)的另一個(gè)重要問(wèn)題是對(duì)工件

14、的總應(yīng)力應(yīng)盡可能均勻分布在定位處。在夾緊過(guò)程中，若P表示反應(yīng)力，那么我們可以確定分散定位接觸力</p><p><b>　?。?）</b></p><p>　　因此,最小化目標(biāo)是確定均衡分散力，最小值（d）。</p><p>　　5. 優(yōu)化夾具設(shè)計(jì)與交換算法</p><p>　　如前所述，在工件表面產(chǎn)生一系列不連續(xù)的拾取

15、點(diǎn)，我們可以為夾具生成可能存在的拾取點(diǎn)。例如,使用CAD模型的工件, 離散的單元平均矢量，可以產(chǎn)生均勻的表面,盡量為夾具零件組件，如圖1。</p><p>　　在夾具布局設(shè)計(jì)中, 從夾具定位的性能要求中，選擇最佳拾取點(diǎn),以及閉合運(yùn)動(dòng)條件。通過(guò)若干個(gè)拾取點(diǎn)，選擇一套合適的定位和夾緊，是一項(xiàng)復(fù)雜的工作。</p><p>　　參考文獻(xiàn)[12～14]已經(jīng)找到理想的有效方法，對(duì)于夾具的質(zhì)量要求在第4

16、部分，是追求最佳方法的一個(gè)交換算法。如圖2的流程圖，其算法由三個(gè)階段組成：</p><p>　　第一階段：隨機(jī)生成初始定位。布局初步隨機(jī)產(chǎn)生選取了包括N個(gè)定位點(diǎn)中的六個(gè)定位點(diǎn)，如果夾緊是預(yù)設(shè)了一套有效的定位，則運(yùn)動(dòng)約束是令人滿意的。反復(fù)產(chǎn)生多種初始定位，每經(jīng)過(guò)進(jìn)一步優(yōu)化的初始定位都會(huì)進(jìn)一步得到最佳結(jié)果。</p><p>　　圖1. 部分CAD模型和收集的夾具拾取點(diǎn)元素</p>

17、<p>　　圖 2. 相互轉(zhuǎn)換算法的流程圖</p><p>　　第二階段：改進(jìn)交換。交換的目標(biāo)是追求一個(gè)注重要求的初步定位設(shè)備?；旧?，這是逐步迭代交換。從目前全面定位與拾取點(diǎn)采集。還必須考慮在閉合形狀約束的交換過(guò)程中。這個(gè)過(guò)程將繼續(xù)改進(jìn)目標(biāo)函數(shù)?；诖鷶?shù)性質(zhì), 一個(gè)高效的最好評(píng)價(jià)，被選為最大目標(biāo)功能（即M），最小△Pc,和最小（d）。</p><p>　　第三階段：選擇最優(yōu)方

18、案。運(yùn)用交換算法隨機(jī)產(chǎn)生一個(gè)初始定位，我們可以為優(yōu)化夾具結(jié)構(gòu)產(chǎn)生幾個(gè)不同的結(jié)果。最佳夾具設(shè)計(jì)能明顯改善最佳目標(biāo)函數(shù)值。應(yīng)當(dāng)強(qiáng)調(diào)，這種算法可應(yīng)用于不同要求的功能。根據(jù)要求, 最好能直接獲得一個(gè)目標(biāo)，或者需要設(shè)計(jì)師的最終決定，來(lái)平衡多重目標(biāo)。</p><p>　　6. 多目標(biāo)優(yōu)化夾具定位</p><p>　　在許多情況下, 通過(guò)最優(yōu)夾具的預(yù)夾緊，再選擇一個(gè)適當(dāng)?shù)亩ㄎ唬湍軜?gòu)建好夾緊區(qū)域,這樣大

19、大提高了功能要求。</p><p>　　采用隨機(jī)算法描述以上內(nèi)容，我們可以分析影響優(yōu)化過(guò)程的特點(diǎn)，為了特殊標(biāo)準(zhǔn)，確定最佳夾具方案。廣泛采用最佳算法夾具設(shè)計(jì)一個(gè)單一要求。根據(jù)統(tǒng)計(jì)和實(shí)例觀察,可得出三個(gè)優(yōu)化完成結(jié)果，最大值（即M），最小值△Pc、（d），一般有些沖突。因此,綜合夾具設(shè)計(jì)需要平衡多重功能要求。文章的重點(diǎn)，進(jìn)一步討論如下。</p><p><b>　　6.1多目標(biāo)均衡&l

20、t;/b></p><p>　　一些定位特性和最小分散力的應(yīng)用是比較重要的。在這種情況下,我們要使用兩步算法：第一，最大值（即M）和第二，最小值（d）。實(shí)驗(yàn)表明，數(shù)值最大化行列式分散力將會(huì)減少，另一方面，分散力減少將導(dǎo)致行列式值下降（即M）。因此, 有人建議平衡兩者之間的具體優(yōu)化目標(biāo)，多目標(biāo)優(yōu)化問(wèn)題的交換算法已成功應(yīng)用于這兩個(gè)要求。當(dāng)夾具預(yù)緊后，閉合形式條件維持每一步交換。</p><p

21、>　　計(jì)算結(jié)果為精確定位的多目標(biāo)優(yōu)化，如圖1、圖3～6。圖3和4為夾具兩步算法進(jìn)行了初步隨機(jī)產(chǎn)生與定位預(yù)夾緊。用最大值(即M)作為第一步換算，已初見(jiàn)成效。當(dāng)行列式增加后，規(guī)范和分散的接觸力將會(huì)減少,可得出一個(gè)全面的高質(zhì)量夾具。首先，多重初始定位設(shè)備以減少優(yōu)化解法交換改進(jìn)。其次，在第二階段應(yīng)用最小值（d），觀察行列式會(huì)減少。兩個(gè)目標(biāo)間的比較，會(huì)產(chǎn)生一個(gè)最終解決方案。這一事件由帕里托線積分來(lái)表示 (圖4)。在這種情況下，由設(shè)計(jì)師最終決

22、定最佳夾具方案。</p><p>　　例如, 最初的定位研究于兩步算法的交換過(guò)程(圖5)進(jìn)一步舉例驗(yàn)證。均衡區(qū)是多目標(biāo)設(shè)計(jì)的關(guān)鍵。每階段后得出的夾具結(jié)構(gòu)，如圖6。應(yīng)當(dāng)指出，第一階段的優(yōu)化最大值（即M）目標(biāo)，定位貼近水平面的部分，且互相遠(yuǎn)離邊界。另一方面，第二階段的優(yōu)化最小值（d）目標(biāo)定位在表面內(nèi)部。</p><p>　　圖3. 變化的特性?shī)A具的兩步優(yōu)化</p><p&g

23、t;　　圖 4. 權(quán)衡期間兩步交換優(yōu)化</p><p>　　圖5.行為兩步交換優(yōu)化設(shè)計(jì)</p><p>　　6.2夾具的最后設(shè)計(jì)決定</p><p>　　在第二階段的算法可能導(dǎo)致行列式值下降。以致于結(jié)果不能解決。為防止這個(gè)問(wèn)題, 交互式控制被設(shè)計(jì)師在最小值（d）換算得到推薦。基本上，由轉(zhuǎn)換過(guò)程來(lái)控制這種行列式（即M）的定位改善必須保持一定約束。然而考慮一定的行列式,

24、 更好的解決方案，是不受優(yōu)化最小值（d）控制。如圖 7 。</p><p>　　圖 6. 夾具配置的兩步優(yōu)化:(a)初始定位；(b)最大值（M）后的定位；(c)最小值（d）后的定位</p><p>　　圖7. 第二階段的兩步換算過(guò)程。左圖：無(wú)約束最小值（d）過(guò)程；右圖：約束最小值（d）過(guò)程</p><p>　　例如, 一套研究在換算過(guò)程的兩種定位控制行列式，見(jiàn) (圖

25、8)。然后列出相應(yīng)的夾具結(jié)構(gòu)圖。這個(gè)例子說(shuō)明對(duì)交互控制的最佳夾具結(jié)構(gòu)性能起決定性作用。</p><p><b>　　7. 優(yōu)化夾具夾緊</b></p><p>　　本節(jié)處理更復(fù)雜的問(wèn)題，同時(shí)為了尋找最佳夾緊和定位，夾具需實(shí)現(xiàn)性能要求。通過(guò)更改夾緊方式，能夠增加夾緊力。這表明，這個(gè)問(wèn)題可以由精確定位目標(biāo)控制。至于其它目標(biāo)，我們必須約束搜索優(yōu)化夾緊內(nèi)部定位點(diǎn)。這樣，優(yōu)化程

26、序能被高效運(yùn)行。</p><p>　　7.1在夾緊中優(yōu)化夾具 </p><p>　　在夾緊應(yīng)用中，有一定的優(yōu)化區(qū)域。因此，有必要從預(yù)選點(diǎn)選擇最佳拾取點(diǎn)。例如，當(dāng)我們覺(jué)得預(yù)選點(diǎn)是優(yōu)化夾具設(shè)計(jì)方面的目標(biāo)精確定位首選點(diǎn)。顯然，如果采用隨機(jī)換算方式，可先后為每個(gè)專(zhuān)用夾具找到最優(yōu)夾具結(jié)構(gòu)。在這些夾具方案中，我們將選擇一個(gè)最佳夾緊定位及其相應(yīng)的夾具優(yōu)化設(shè)計(jì)。如圖9和圖10。 </p>&

27、lt;p>　　圖8.均衡兩步優(yōu)化 (a)兩個(gè)目標(biāo)之間均衡</p><p>　　(b)高出B1的行列式夾具配置情況(c)高出B2的行列式夾具配置情況</p><p>　　圖9. 從單目標(biāo)設(shè)計(jì)中選擇夾緊</p><p>　　圖10. 初始夾具拾取點(diǎn)(左)以及相應(yīng)的最佳夾具和相應(yīng)的定位點(diǎn)(右)</p><p>　　圖11. 多目標(biāo)夾具設(shè)計(jì)的

28、選擇</p><p>　　圖12. 初始夾緊選擇（左）和最佳夾緊選擇及相對(duì)應(yīng)的優(yōu)化定位</p><p>　　7.2從一套夾具中優(yōu)化夾具</p><p>　　另外,優(yōu)化夾具設(shè)計(jì)可以延伸到多目標(biāo)設(shè)計(jì)問(wèn)題。在每個(gè)夾具中主要應(yīng)用兩步換算方法。</p><p>　　在應(yīng)用所有夾具選擇這一結(jié)果的過(guò)程中，我們可以比較結(jié)果，來(lái)選擇最適合夾具。如上所述，在考慮

29、多方面要求均衡，每個(gè)夾具會(huì)產(chǎn)生不同的結(jié)果。這表現(xiàn)在圖11和圖10。最佳夾緊應(yīng)定位精確和接觸力均衡，如圖11。最后設(shè)計(jì)最優(yōu)的相應(yīng)夾具定位，見(jiàn)圖12，當(dāng)夾緊定位在左邊。</p><p><b>　　8. 結(jié)論</b></p><p>　　本文針對(duì)三維工件的夾具布局優(yōu)化設(shè)計(jì)轉(zhuǎn)換算法。多目標(biāo)優(yōu)化的目標(biāo)包括工件定位準(zhǔn)確，較少力和平衡接觸力。本文著重于多基準(zhǔn)優(yōu)化設(shè)計(jì)與分層研究。利

30、用優(yōu)化工藝過(guò)程，以實(shí)現(xiàn)高效率轉(zhuǎn)換均衡各指定目標(biāo)。并實(shí)例來(lái)說(shuō)明設(shè)計(jì)過(guò)程及其成效。</p><p>　　定位和夾緊的相互關(guān)系，在夾具質(zhì)量測(cè)量中起決定性作用，在今后的優(yōu)化夾具定位中需要更為連貫和完整的研究方法。</p><p><b>　　9.參考文獻(xiàn)</b></p><p>　　[1] Campbell PD. Basic fixture desi

31、gn. New York: Industrial Press,</p><p><b>　　1994.</b></p><p>　　[2] Reuleaux F. The kinematics of machinery. New York: Dover</p><p>　　Publications, 1963.</p><p

32、>　　[3] Mishra B, Schwartz JT, Sharir M. On the existence and</p><p>　　synthesis of multifinger positive grips. Algorithmica 1987;</p><p>　　2(4):541–58.</p><p>　　[4] Markenscoff X

33、, Ni L, Papadimitriou CH. The geometry of</p><p>　　grasping. Int J Robot Res 1990;9(1):61–74.</p><p>　　[5] Chou Y-C, Chandru V, Barash MM. A mathematicalapproach</p><p>　　to automat

34、ed configuration of machining fixtures: analysis and</p><p>　　synthesis. J Eng Ind 1989;111:299–306.</p><p>　　[6] DeMeter EC. Restraint analysis of fixtures which rely on surface</p><

35、p>　　contact. J Eng Ind 1994;116(2):207–15.</p><p>　　[7] Asada H, By AB. Kinematics analysis of workpart fixturing for</p><p>　　flexible assembly with automatically reconfigurable fixtures. IE

36、EE</p><p>　　J Robot Automat 1985;RA1:86–93.</p><p>　　[8] Brost RC, Goldberg KY. A complete algorithm for designing</p><p>　　planar fixtures using modular components. IEEE Trans Robo

37、t</p><p>　　Automat 1996;12(1):31–46.</p><p>　　[9] Zhuang Y, Goldberg K, Wong Y-C. On the existence of solutions</p><p>　　in modular fixturing. Int J Robot Res 1996;15(5):5–9.</p&

38、gt;<p>　　[10] Cai W, Hu SJ, Yuan J. A variationalmethod of robust fixture</p><p>　　configuration design for 3-d workpieces. J Manuf Sci Eng</p><p>　　1997;119:593–602.</p><p>

39、　　[11] Baraff D, Mattikalli R, Khosla P.Minimal fixturing of frictionless</p><p>　　assemblies: complexity and algorithms. Algorithmica 1997;19:</p><p><b>　　4–39.</b></p><p

40、>　　[12] Wang MY. An optimaldesign for 3D fixture synthesis in a</p><p>　　point set domain. IEEE Trans Robot Automat 2000;16(6):</p><p><b>　　833–41.</b></p><p>　　[13]

41、Atkinson A, Doney A. Optimum experimentaldesigns.</p><p>　　New York: Oxford University Press, 1992.</p><p>　　[14] Wang MY. Optimizing fixture layout in a point set domain. IEEE</p><p&

42、gt;　　Trans Robot Automat 2001;17(3):312–23.</p><p>　　Flexible and reconfigurable manufacturing systems paradigms</p><p>　　Hoda A. ElMaraghy</p><p>　　Abstract Reconfigurable Manufact

43、uring System (RMS) is a new manufacturing systems paradigm that aims at achieving cost-effective and rapid system changes, as needed and when needed, by incorporating principles of modularity, integrability, flexibility,

44、 scalability, convertibility, and diagnosability. RMS promises customized flexibility on demand in a short time, while Flexible Manufacturing Systems (FMSs) provides generalized flexibility designed for the anticipated v

45、ariations and built-in a pri</p><p>　　of changeability and transformability of manufacturing systems, their components as well as factories, are presented along with their enablers and compared with flexibil

46、ity and reconfigurability. The importance of having harmonized human-machine manufacturing systems is highlighted and the role of people in the various manufacturing paradigms and how this varies in pursuit of productivi

47、ty are illustrated. Finally, the</p><p>　　industrial and research challenges presented by these manufacturing paradigms are discussed.</p><p>　　Keywords Changeability · Flexibility · M

48、anufacturing systems · Reconfiguration</p><p>　　1. Introduction</p><p>　　Manufacturing systems have evolved from job shops, which feature general-purpose machines, low variety dedicated man

49、ufacturing lines driven by the economy of scale. In the eighties the concept of flexible manufacturing was introduced in response to the need for mass customization and for greater responsiveness to changes in products,&

50、lt;/p><p>　　production technology, and markets. Flexible manufacturing systems were also developed to address mid-volume, mid-variety production needs. Similarities between parts in design and/or manufacture we

51、re used to achieve economy of scope. Flexible manufacturing systems (FMSs) anticipated these variations and built-in flexibility a priori; hence they are more robust but have high initial capital investment cost. The<

52、/p><p>　　flexibility attributes are sometimes underused. In the nineties, optimality, agility,waste reduction, quality, and lean manufacturing were identified as key drivers and goals for ensuring survival in a

53、 globally competitive market.</p><p>　　The reconfigurable manufacturing concept has emerged in the last few years in an attempt to achieve changeable functionality and scalable capacity (Koren et al., 1999;

54、Fujii et al., 2000). It proposes a manufacturing system where machine components, machines, cells, or material handling units can be added, removed, modified,</p><p>　　or interchanged as needed to respond qu

55、ickly to changing requirements. Such a fully reconfigurable system does not yet exist today but is the subject of major research efforts around the world, with special emphasis on the hardware and machine control aspects

56、. Proponents of this approach believe that it has the potential to offer a cheaper solution, in the long run, compared to FMSs, as it can increase the life and utility of a</p><p>　　manufacturing system. Har

57、dware reconfiguration also requires major changes in the software used to control individual machines, complete cells, and systems as well as to plan and control the individual processes and production. All this adds to

58、the evergrowing complexity of products, processes, manufacturing systems, and enterprises (Wiendahl and Scholtissek, 1994).</p><p>　　In this paper, the highlights of recent research into the notion of manufa

59、cturing system flexibility and its measurement and impact are reviewed. Various types/classification of flexibilities are presented with a view to clarify their correspondence with some aspects of manufacturing systems r

60、econfiguration. The characteristics and pre-requisites of a reconfigurable manufacturing system are overviewed. The concept of a manufacturing system life cycle is introduced and linked with aspects of manu</p>&l

61、t;p>　　2. Manufacturing systems flexibility</p><p>　　Flexibility attracted much attention from researchers to better understand and clarify its concept. As a result, many definitions emerged in the literat

62、ure. Early definitions related to the flexibility of manufacturing systems are based on the notion of adaptability to uncertainties (Mandelbaum, 1978; Slack, 1987). </p><p>　　2.1. Manufacturing flexibility c

63、lassification</p><p>　　Review of the literature identifies at least 10 types of manufacturing systems flexibilities</p><p>　　(see Browne et al., 1984; and Sethi and Sethi, 1990). These are:</

64、p><p>　　1. Machine flexibility: Various operations performed without set-up change,</p><p>　　2. Material handling flexibility: Number of used paths / total number of possible paths between all mach

65、ines,</p><p>　　3. Operation Flexibility: Number of different processing plans available for part fabrication,</p><p>　　4. Process Flexibility: Set of part types that can be produced without majo

66、r set-up changes, i.e. part-mix flexibility,</p><p>　　5. Product Flexibility: Ease (time and cost) of introducing products into an existing product mix. It contributes to agility,</p><p>　　6. Ro

67、uting Flexibility: Number of feasible routes of all part types/Number of part types,</p><p>　　7. Volume Flexibility: The ability to vary production volume profitably within production capacity,</p>&l

68、t;p>　　8. Expansion Flexibility: Ease (effort and cost) of augmenting capacity and/or capability, when needed, through physical changes to the system,</p><p>　　9. Control Program Flexibility: The ability o

69、f a system to run virtually uninterrupted (e.g. during the second and third shifts) due to the availability of intelligent machinesand system control software,</p><p>　　10. Production Flexibility: Number of

70、all part types that can be produced without adding major capital equipment.</p><p>　　This classification promotes better understanding of various types of flexibility although</p><p>　　some of t

71、hem are inter-related. It should be noted that the expansion flexibility is</p><p>　　consistent with the current understanding of manufacturing systems reconfigurability.</p><p>　　3. Manufacturi

72、ng systems life cycle</p><p>　　The significant reduction in product development time brought about by the use of CADtoolswas not paralleled in the design and development of manufacturing systems. These syste

73、ms must be designed to satisfy certain requirements and constraints that vary over time. Recent improvements in productivity were attributed more to improvements in the design and operation of manufacturing systems, as w

74、ell as the design of products, than to manufacturing processes or technology improvements. Some modern d</p><p>　　4. Manufacturing system reconfiguration</p><p>　　The changing manufacturing envi

75、ronment characterized by aggressive competition on a global scale and rapid changes in process technology requires careful attention to prolonging the life of manufacturing systems by making them easily up-gradable and i

76、nto which new technologies and new functions can be readily integrated. A reconfigurable manufacturing system (RMS) is a visionary challenge for manufacturing</p><p>　　enterprises and is considered as the ne

77、xt manufacturing paradigm (Agility Forum, 1997 and NRC (U.S.), 1998). They would use modular equipment as building blocks to realize the required system functionality for the production of a part family. Instead of provi

78、ding a general flexibility through the use of equipment with built-in high functionality, as in FMSs, RMSs provide customized flexibility through scalability and reconfiguration as needed when needed to meet market requi

79、rements (Mehrabi et a</p><p><b>　　2000). </b></p><p>　　5. Flexibility and reconfigurability</p><p>　　The design, characteristics, and potential merits of RMSs and how th

80、ey compare with other manufacturing paradigms have been occupying researchers and practitioners at this stage of manufacturing systems evolution. A panel of experts from industry and academia was assembled recently to di

81、scuss and debate these issues. They were asked to address the similarities and differences between RMS and FMS, the definitions</p><p>　　of flexibility, reconfigurability, and changeability, and how to chara

82、cterize a manufacturing system’s responsiveness. These opinions are summarized in the following sections.</p><p>　　5.1. Flexible and reconfigurable manufacturing system</p><p>　　An RMS is design

83、ed at the outset for a possible rapid change in structure, as well as in hardware and software components, in order to quickly adjust production capacity and functionality within a part family. An FMS is a system whose m

84、achines are able to perform operations on a random sequence of parts of different types with little or no time or other expenditure for changeover. In practice, FMSs consist of processing stations and material handling s

85、ystems that are entirely under computer cont</p><p>　　5.2. RMS key characteristics</p><p>　　The key characteristics of RMS, including modularity, integrability, flexibility, scalability, convert

86、ibility, and diagnosability, were emphasized as prerequisites to enable reconfigurable manufacturing systems to work as intended and achieve the desired reduction in time and cost . The concept of reconfigurable machine

87、tools</p><p>　　(RMTs), which are designed with customized flexibility that enable combining the advantages of high productivity of dedicated stations with the flexibility of CNCs was presented and compared a

88、s summarized . These are illustrated with an example of a proposed design of a reconfigurable machine tool (RMT) with reconfigurable spindle heads developed at the NSF Engineering Research Center for Reconfigurable Manuf

89、acturing Systems at University of Michigan, Ann Arbor (Kota and Koren, 1999; Landers et </p><p>　　5.3. Flexibility is the future of reconfigurability</p><p>　　It can be noted that there are suff

90、icient common grounds in philosophy and application between the FMS and RMS paradigms to support the notion that they represent a continuum and to predict that: “the Future Reconfigurable Manufacturing Systems will be Mo

91、re Flexible” (Stecke, 2005). Some manufacturing support functions and intelligent software will be necessary to achieve effective reconfiguration, such as:</p><p>　　1) Software that can help select the best

92、equipment (machine tools) based on their capabilities, accuracies, and flexibilities and the best materials, tooling, coolant, fixtures/ fixturing to be used to machine any particular component, 2) Future CAD/CAM will re

93、cognize geometrical shapes in downloaded customer CAD files and automatically generateCNCpart programs that will include appropriate speeds, feed rates, tools</p><p>　　and fixtures selections, and process/op

94、eration sequencing, 3) Tooling, fixturing, gauging equipment, and supplies could be automatically retrieved from inventory stores in an ASRS and introduced into the process at appropriate times, and 4) The process of eac

95、h ID coded part could be automatically tracked through the various manufacturing stages, allowing manufacturing and cost analysis at every stage of production.</p><p>　　5.4. Reconfiguration with available te

96、chnology</p><p>　　The machine tool industry, represented by one of its successful players, voiced an opinion that there is a compelling value proposition for reconfiguration, the only qualification being cos

97、t and availability (Hyatt, 2005). It is clear that reconfigurable machine tools (RMTs) are an essential enabler of RMSs. However, the current stateof- the-art is such that broadly reconfigurable machine tools are not yet

98、 available as</p><p>　　the required technology is still in various states of development. It was suggested that there are ways to achieve many of the potential benefits of reconfigurable manufacturing system

99、s, and make the users of machine tools and manufacturing systems more profitable while this new technology is under development. For example, it is possible to replace non-reconfigurable machine tools (NRMTs) by machines

100、 of alternate configurations provided that certain system features are implemented to facilitate</p><p>　　5.5. Changeability, reconfigurability and flexibility of manufacturing systems</p><p>　　

101、The concept of having a flexible, reconfigurable and changeable factory infrastructure to support the re-deployment of machines and reconfiguration of systems discussed above resonated strongly with the notion of changea

102、bility, including that manifested in plant physical structures and buildings. Changeability of manufacturing systems has been a focus of discussion and analysis within the International Academy for</p><p>　　

103、Production Engineering (CIRP) academic and industrial community for a number of years (Wiendahl, 2003, 2005). It resulted in a classification of changeability and its drivers and enablers as well as its relationship with

104、 flexibility and reconfiguration, as shown in Figs. 4 and 5. It is important to assess the degree of changeability of current and planned factories, given some prevailing major trends in industry including: (</p>