工業(yè)工程外文翻譯--裝配線平衡問(wèn)題的求解方法淺析

1、<p>　　畢業(yè)設(shè)計(jì)(論文)英文翻譯</p><p>　　學(xué)生姓名： 蔣偉偉 學(xué) 號(hào)： 2203080128 </p><p>　　所在學(xué)院： 經(jīng)濟(jì)與管理學(xué)院 </p><p>　　專 業(yè)： 工業(yè)工程 <

2、/p><p>　　設(shè)計(jì)(論文)題目：好孩子推車(chē)事業(yè)部總裝線生產(chǎn)線優(yōu)化設(shè)計(jì)</p><p>　　指導(dǎo)教師： 蔣南云 </p><p>　　2012年4月25號(hào)</p><p>　　The Methods for Solving the Problem of Balancing an

3、 Assembly Line</p><p>　　Currently, along with the market changing, some forerunners production mode got an extensive application in the manufacturing industry. How raise a whole efficiency of assembling the

4、production line, reduce a work preface in the ware, and pursue to synchronize production is valued by more and more people. The production of manufacturing industry is most likely after carrying on subdividing to turn of

5、 have another a work preface flowing water to turn a continuous homework production line, at this</p><p>　　The assemble is the last link of production, assembling process mainly with the gearing of parts, ti

6、ghtly solid in lord; secondly allied connect, press to pack and add to note various work to lie quality and quality examination of work preface, sometimes still want to choose to pack according to the customer intention.

7、 The whole assemble homework is complicated, belonging to a labor an intensive type engineering. Therefore, the balance of exaltation assembly line has important realistic meaning </p><p>　　The assembly line

8、 equilibrium problem is a typical NP-hard to crack, currently, the method which solves this kind of problem can return to knot to 3: Inspire type method, superior turn method (mathematics model method) with industrial en

9、gineering method.　　Superior turn method is the superior solution that passes establishment mathematics model to look for a problem. Though the mathematics model method can find out the superior solution, the physically

10、applied hour is very tedious; Usually a pim</p><p>　　The linear programming (Linear Programming, brief name LP) problem is a target function peace treaty's tying conditions are the all lines and superior

11、 to turn a problem. The line programming is the superior to turn one of the important categories in the problem realm, making use of the line programming method to solve an ALB problem is in 60s in 20 centuries waited fo

12、r someone by Bowman EH to put forward, its stipulation condition and target function establishment easy, but the operation measure</p><p>　　The dynamic plan (dynamic programming) the algorithm is one commonl

13、y used method that solves the multi- stages decision-making process optimization question; it is very difficult and skillful. The dynamic plan realization is not certainly complex, is suitable for many questions, when so

14、lution general question is one of we first choice algorithms. But, the dynamic plan mathematical model establishment is not an easy matter, most difficult also most importantly condition expression. The dynamic pl</p&

15、gt;<p>　　The heuristic method by its simple, easy to understand, was fast and has satisfied won the multitudinous administrative personnel's trust and the welcome, and is widely applied to each domain. Its pro

16、duction mainly is for overcome the realistic modeling the difficulty, provides one kind of more effective policy-making tool Compares with the optimized method, the heuristic method merit mainly has: The logical model ap

17、proaches to the reality, the flow chart establishment in the foundation which th</p><p>　　The heredity algorithm (genetic algorithms, was called GA) is J. Holland was proposed in 1975 the biological theory o

18、f evolution inspiration. GA is based on the survival of the fittest one kind highly parallel, stochastic and the auto-adapted optimized algorithm, he expressed the question solution chromosome survival of the fittest pro

19、cess, through the group generation of unceasingly evolves, including duplication, overlapping and operation and so on variation, finally restrains to the individua</p><p>　　The industrial engineering (Indust

20、rial Engineering, is called IE) the thought most early is proposes father of Taylor by scientific management in the beginning of 20th century, it is the integrated system which by the person, the material, the equipment

21、or the facility, the energy and the information composes for the main research object, synthesis application project technology, management science and social sciences knowledge and so on theory method, enable sit to ach

22、ieve reduces the cost, imp</p><p>　　The assembly line balance question has experienced from the optimize algorithm to the heuristic algorithm, until modern artificial intelligence method and industrial engin

23、eering method process. This article in view of the linear programming law, the dynamic plan algorithm, the heredity algorithm and the industrial engineering method these four solutions assembly line balance question typi

24、cal method has separately made the brief analysis, has pointed out each algorithm good and bad points and the</p><p>　　裝配線平衡問(wèn)題的求解方法淺析</p><p>　　目前，隨著快速多變的市場(chǎng)需求，一些先進(jìn)的生產(chǎn)模式在制造業(yè)得到了廣泛的應(yīng)用。如何提高裝配生產(chǎn)線的整體效率

25、，減少工序間的在制品，以及追求同步化生產(chǎn)越來(lái)越受到人們的重視。制造業(yè)的生產(chǎn)多半是在進(jìn)行細(xì)分化之后的多工序、流水化、連續(xù)作業(yè)生產(chǎn)線，此時(shí)由于分工作業(yè)，各工序的作業(yè)時(shí)間在理論上、實(shí)際操作上都不能完全相同，這就勢(shì)必存在工序間作業(yè)負(fù)荷不均的現(xiàn)象。除了造成無(wú)謂的工時(shí)損失外，還造成大量的工序堆積，嚴(yán)重時(shí)會(huì)造成生產(chǎn)線的中止。裝配線平衡(Assembly Line Balancing，簡(jiǎn)稱ALB)就是為了解決上述問(wèn)題而提出的一種手段與方法，它對(duì)各工序的

26、作業(yè)時(shí)間進(jìn)行平均化，同時(shí)對(duì)作業(yè)進(jìn)行研究，對(duì)時(shí)間進(jìn)行測(cè)定，使裝配生產(chǎn)線順暢連動(dòng)。</p><p>　　裝配是生產(chǎn)的最后一個(gè)環(huán)節(jié)，裝配過(guò)程主要是以零部件的安裝、緊固為主；其次是聯(lián)接、壓裝和加注各種工作介質(zhì)以及質(zhì)量檢測(cè)的工序，有時(shí)還要根據(jù)用戶意向選裝。整個(gè)裝配作業(yè)繁雜，屬勞動(dòng)密集型工程。因此，提高裝配線的平衡對(duì)于提高汽車(chē)裝配線的生產(chǎn)效率有著重要的現(xiàn)實(shí)意義。裝配線平衡問(wèn)題是一類(lèi)典型的離散型組合優(yōu)化問(wèn)題，尤其是對(duì)于隨機(jī)的、

27、多目標(biāo)的裝配線平衡問(wèn)題，在某種程度上較難得到滿意解。目前，對(duì)于裝配線平衡問(wèn)題的研究主要分為以下3個(gè)方面：給定裝配線的節(jié)拍，求最小工作站數(shù)，通常在裝配線的設(shè)計(jì)與安裝階段進(jìn)行；給定裝配線的最小工作站數(shù)，使裝配線的節(jié)拍最小，對(duì)已存在的生產(chǎn)線進(jìn)行調(diào)整優(yōu)化；在裝配線的工作站數(shù)和節(jié)拍得到優(yōu)化確定的條件下，均稀裝配線上工作站的負(fù)荷，給工作人員一種公平感。由于裝配線的平衡總是離散型組合優(yōu)化問(wèn)題，尋求最優(yōu)解比較困難，故采取合適的方法解決裝配線平衡問(wèn)題引起

28、了各企業(yè)界和學(xué)術(shù)界的廣泛關(guān)注。本文旨對(duì)流水線平衡問(wèn)題的幾種典型求解方法進(jìn)行比較分析，提出解決該問(wèn)題可行方法。</p><p>　　裝配線平衡問(wèn)題是典型的NP-hard難題，目前，求解此類(lèi)問(wèn)題的方法可歸結(jié)為3種類(lèi)型：?jiǎn)l(fā)式方法、最優(yōu)化方法(數(shù)學(xué)模型法)和工業(yè)工程方法。</p><p>　　最優(yōu)化方法就是通過(guò)建立數(shù)學(xué)模型來(lái)尋找問(wèn)題的最優(yōu)解。盡管數(shù)學(xué)模型法能找到最優(yōu)解，但實(shí)際應(yīng)用時(shí)十分繁瑣；往往

29、一個(gè)很小的問(wèn)題需要構(gòu)造的模型非常大，計(jì)算機(jī)耗時(shí)也較多。下面分別就線性規(guī)劃法和動(dòng)態(tài)規(guī)劃法作一簡(jiǎn)述與分析。</p><p>　　線性規(guī)劃(Linear Programming，簡(jiǎn)稱LP)問(wèn)題是目標(biāo)函數(shù)和約束條件都是線性的最優(yōu)化問(wèn)題。線性規(guī)劃是最優(yōu)化問(wèn)題領(lǐng)域中重要的范疇之一，利用線性規(guī)劃方法求解ALB問(wèn)題是在20世紀(jì)60年代由Bowman EH等人提出，其約束條件和目標(biāo)函數(shù)建立容易，但運(yùn)算量很大，與實(shí)際情況相差比較遠(yuǎn)。

30、之后一些學(xué)者將此算法加以改進(jìn)，但運(yùn)算量也比較大。由于運(yùn)用線性規(guī)劃模型方法求解運(yùn)算量很大，只能求解作業(yè)元素個(gè)數(shù)較少的線平衡問(wèn)題，與實(shí)際情況相差甚大，故此法在實(shí)際應(yīng)用中所取得的效果不是很理想。</p><p>　　動(dòng)態(tài)規(guī)劃(dynamic programming)算法是解決多階段決策過(guò)程最優(yōu)化問(wèn)題的一種常用方法，難度比較大，技巧性也很強(qiáng)。動(dòng)態(tài)規(guī)劃實(shí)現(xiàn)并不復(fù)雜，適用于許多問(wèn)題，在解決一般問(wèn)題時(shí)是我們首選的算法之一。但是

31、，動(dòng)態(tài)規(guī)劃的數(shù)學(xué)模型的建立不是件容易的事，其中最困難也最重要的是狀態(tài)表示。動(dòng)態(tài)規(guī)劃的狀態(tài)表示描述的子問(wèn)題必須滿足最優(yōu)子結(jié)構(gòu)性質(zhì)，否則無(wú)法建立正確的動(dòng)態(tài)規(guī)劃模型。在應(yīng)用動(dòng)態(tài)規(guī)劃方法解決問(wèn)題時(shí)，應(yīng)先估計(jì)問(wèn)題的時(shí)間、空間，如果問(wèn)題存在維數(shù)障礙，那么動(dòng)態(tài)規(guī)劃的狀態(tài)表示很難滿足較大規(guī)模問(wèn)題的空間要求。我們必須另尋其他方法。動(dòng)態(tài)規(guī)劃法運(yùn)用于解決ALB問(wèn)題時(shí)，其狀態(tài)的表示與各工作站作業(yè)元素的加工時(shí)間有關(guān)，往往在確定動(dòng)態(tài)上界時(shí)需對(duì)系統(tǒng)影響較小的狀態(tài)進(jìn)行

32、刪除，從而達(dá)到降低運(yùn)算量的目的。</p><p>　　啟發(fā)式方法以其簡(jiǎn)便、易懂、快速和滿意贏得了眾多管理人員的信賴和歡迎,并被廣泛應(yīng)用于各個(gè)領(lǐng)域。它的產(chǎn)生主要是為了克服現(xiàn)實(shí)建模的困難，提供一種更有效的決策工具。與最優(yōu)化方法相比較，啟發(fā)式方法的優(yōu)點(diǎn)主要有：邏輯模型接近于現(xiàn)實(shí)，流程圖建立在決策者經(jīng)驗(yàn)的基礎(chǔ)上，因此啟發(fā)式方法隱含著多目標(biāo)方案；它的主要局限是它的靜態(tài)性，即在平衡過(guò)程中，固定的準(zhǔn)則是預(yù)先確定的優(yōu)先準(zhǔn)則，而從

33、產(chǎn)品加工過(guò)程、市場(chǎng)需求和公司戰(zhàn)略來(lái)看，環(huán)境是動(dòng)態(tài)變化的。</p><p>　　遺傳算法(genetic algorithms，簡(jiǎn)稱GA)是J. Holland于1975年受生物進(jìn)化論的啟發(fā)而提出的。GA是基于適者生存的一種高度并行、隨機(jī)和自適應(yīng)的優(yōu)化算法，他將問(wèn)題的求解表示成染色體的適者生存過(guò)程，通過(guò)染色體群的一代代不斷進(jìn)化，包括復(fù)制、交叉和變異等操作，最終收斂到最適應(yīng)環(huán)境的個(gè)體，從而求得問(wèn)題的最優(yōu)解或滿意解。與

34、傳統(tǒng)優(yōu)化方法相比，遺傳算法的優(yōu)點(diǎn)是：群體搜索；不需要目標(biāo)函數(shù)的導(dǎo)數(shù)；概率轉(zhuǎn)移準(zhǔn)則。近年來(lái)，由于遺傳算法求解復(fù)雜優(yōu)化問(wèn)題的巨大潛力及其在工業(yè)工程領(lǐng)域的成功應(yīng)用，這種算法得到了廣泛的關(guān)注。遺傳算法在實(shí)際的應(yīng)用中往往出現(xiàn)早熟收斂和收斂性能差等缺點(diǎn)，現(xiàn)今的一些改進(jìn)方法大都是針對(duì)基因操作、種群的宏觀操作、基于知識(shí)的操作和并行化GA進(jìn)行。王云鳳等人運(yùn)用混合遺傳算法求解裝配線平衡問(wèn)題，該算法將啟發(fā)式規(guī)則、遺傳算法、模擬退火算法綜合在一起，既利用了啟發(fā)

35、式規(guī)則基于經(jīng)驗(yàn)的優(yōu)勢(shì)，可以提高算法的速度，又可使遺傳算法和模擬退火算法優(yōu)勢(shì)互補(bǔ)，有效地解決了裝配線平衡問(wèn)題。</p><p>　　工業(yè)工程(Industrial Engineering，簡(jiǎn)稱IE)思想最早是由科學(xué)管理之父泰勒在20世紀(jì)初提出的，它是以人、物料、設(shè)備或設(shè)施、能源和信息組成的集成系統(tǒng)為主要研究對(duì)象，綜合應(yīng)用工程技術(shù)、管理科學(xué)和社會(huì)科學(xué)的理論方法等知識(shí)，使其達(dá)到降低成本、提高質(zhì)量和效益的目的，同時(shí)為科學(xué)

36、管理提供決策依據(jù)。以工業(yè)工程為主的生產(chǎn)線平衡相關(guān)技術(shù)主要是方法研究和作業(yè)測(cè)定兩大技術(shù)，運(yùn)用該技術(shù)的優(yōu)點(diǎn)就是使企業(yè)在不投資或少投資情況下，不增加工人勞動(dòng)強(qiáng)度甚至是降低勞動(dòng)強(qiáng)度，通過(guò)實(shí)施一系列適合自身特點(diǎn)改善方法，對(duì)生產(chǎn)過(guò)程的作業(yè)程序、作業(yè)法、MOD法、物料配置、空間布局及作業(yè)環(huán)境等各方面進(jìn)行改善，達(dá)到企業(yè)平衡生產(chǎn)線進(jìn)而提高生產(chǎn)能力、取得經(jīng)濟(jì)效益的目標(biāo)。改善是企業(yè)永恒的主題，而工業(yè)工程所追求的目標(biāo)就是永無(wú)止境的改善，運(yùn)用相關(guān)技術(shù)對(duì)生產(chǎn)線平衡

37、，為企業(yè)降低消耗，提高生產(chǎn)率。最近一、二十年，工業(yè)的內(nèi)外環(huán)境發(fā)生了很大的變化，知識(shí)經(jīng)濟(jì)時(shí)代正在到來(lái)，生產(chǎn)線平衡方法體系也將具有新的特點(diǎn)和研究方向，以適應(yīng)新環(huán)境下經(jīng)濟(jì)增長(zhǎng)的需要。此外還有設(shè)備布置等相關(guān)方法也可以適當(dāng)提高生產(chǎn)線平衡率，為企業(yè)帶來(lái)更大的經(jīng)濟(jì)效益，有待我們繼續(xù)做進(jìn)一</p><p>　　裝配線平衡問(wèn)題經(jīng)歷了一個(gè)由最優(yōu)化算法到啟發(fā)式算法，直到現(xiàn)代人工智能方法和工業(yè)工程方法的過(guò)程。本文針對(duì)線性規(guī)劃法、動(dòng)態(tài)規(guī)劃

38、算法、遺傳算法和工業(yè)工程方法這四種求解裝配線平衡問(wèn)題的典型方法分別作了簡(jiǎn)要分析，指出了各種算法的優(yōu)缺點(diǎn)及應(yīng)用情況。裝配線平衡問(wèn)題是一類(lèi)典型的離散型組合優(yōu)化問(wèn)題，尤其是對(duì)于隨機(jī)的、多目標(biāo)的裝配線平衡問(wèn)題，在某種程度上采用合適的方法都能得到滿意解。在應(yīng)用優(yōu)化方法求解的同時(shí),幾乎都用到了仿真技術(shù)，隨著計(jì)算機(jī)技術(shù)的發(fā)展，在目前現(xiàn)有軟件包的基礎(chǔ)上進(jìn)行二次開(kāi)發(fā)，建立可視化的虛擬仿真裝配系統(tǒng)，以實(shí)現(xiàn)裝配線的動(dòng)態(tài)平衡效果，將值得進(jìn)一步研究和探討。目前，