梁艷華
關(guān)鍵詞: 大數(shù)據(jù); 智能管理; 企業(yè)管理系統(tǒng); 數(shù)據(jù)處理; 故障預(yù)測(cè); 冗余信息
中圖分類號(hào): TN919?34; F272 ? ? ? ? ? ? ? ? ? 文獻(xiàn)標(biāo)識(shí)碼: A ? ? ? ? ? ? ? ? ? ?文章編號(hào): 1004?373X(2019)06?0158?04
Abstract: Since the traditional enterprise management system built in the network environment has a poor effect on big data analysis in enterprise management and cannot improve the efficiency of enterprise operation and management, an intelligent enterprise management system based on big data is designed. The management and application innovation module is adopted to realize enterprise intelligent management and application innovation by means of multi?screen interaction and other modes. The big data processing module is used to collect, transmit and store data. The three?layer architecture is used for the system software to design the data compilation process, so as to eliminate related information. The redundant information is eliminated using the data cleaning process. The fault prediction method based on the big data behavior model is used to analyze time series, so as to realize fault pre?judgment of the system. The experimental results show that the expected results of the system test are consistent with the actual results, the maximum throughput of the system is 290 000 Mb/s, and its average response time is 3.5 s, which indicates that the designed system can achieve efficient intelligent management of enterprises.
Keywords: big data; intelligent management; enterprise management system; data processing; fault prediction; redundant information
大數(shù)據(jù)技術(shù)已經(jīng)成為企業(yè)發(fā)展出奇制勝的關(guān)鍵點(diǎn),伴隨數(shù)據(jù)存儲(chǔ)成本的減少與移動(dòng)互聯(lián)網(wǎng)的普及,大數(shù)據(jù)技術(shù)運(yùn)用在企業(yè)大事件的分析與預(yù)測(cè)中。文獻(xiàn)[1]通過處理Web網(wǎng)絡(luò)數(shù)據(jù)分類問題,實(shí)現(xiàn)企業(yè)數(shù)據(jù)的智能管理,但是該方法未在企業(yè)管理中廣泛應(yīng)用大數(shù)據(jù)管理方法,導(dǎo)致企業(yè)管理效率低;文獻(xiàn)[2]對(duì)網(wǎng)絡(luò)環(huán)境下的企業(yè)管理系統(tǒng)進(jìn)行設(shè)計(jì),對(duì)企業(yè)管理中的大數(shù)據(jù)分析效果差,存在響應(yīng)時(shí)間長(zhǎng)的弊端;文獻(xiàn)[3]構(gòu)建了基于大數(shù)據(jù)集群架構(gòu)的數(shù)據(jù)管理系統(tǒng),該系統(tǒng)缺乏對(duì)大數(shù)據(jù)中冗余信息的清洗過程,不能準(zhǔn)確判斷出系統(tǒng)中的故障數(shù)據(jù),管理性能差。為了解決上述問題,本文設(shè)計(jì)基于大數(shù)據(jù)的智能企業(yè)管理系統(tǒng),實(shí)現(xiàn)高效率的企業(yè)智能管理。
1.1 ?系統(tǒng)總體結(jié)構(gòu)設(shè)計(jì)
設(shè)計(jì)的基于大數(shù)據(jù)的智能企業(yè)管理系統(tǒng)的結(jié)構(gòu)如圖1所示。大數(shù)據(jù)環(huán)境下,企業(yè)群體智能維度與人機(jī)結(jié)合智能維度構(gòu)成企業(yè)智能管理系統(tǒng)框架結(jié)構(gòu)。
1.2 ?管理應(yīng)用創(chuàng)新模塊結(jié)構(gòu)設(shè)計(jì)
圖2為管理應(yīng)用創(chuàng)新模塊結(jié)構(gòu)圖。分析圖2可得,產(chǎn)品數(shù)據(jù)與用戶行為數(shù)據(jù)表示客戶洞察維度、產(chǎn)品設(shè)計(jì)維度、精準(zhǔn)營(yíng)銷維度等,這些維度是企業(yè)智能管理系統(tǒng)應(yīng)用的創(chuàng)新內(nèi)容。為完成客戶的細(xì)查,使用大規(guī)模數(shù)據(jù)和實(shí)時(shí)性數(shù)據(jù)可縮短企業(yè)收集用戶數(shù)據(jù)的時(shí)間,使企業(yè)經(jīng)營(yíng)管理效率與用戶體驗(yàn)大幅度提升[4]。

1.3 ?大數(shù)據(jù)處理模塊設(shè)計(jì)
數(shù)據(jù)采集、數(shù)據(jù)導(dǎo)入與預(yù)處理、數(shù)據(jù)存儲(chǔ)、數(shù)據(jù)處理與結(jié)果展現(xiàn)是系統(tǒng)大數(shù)據(jù)的處理流程[5]。圖3是大數(shù)據(jù)處理模塊結(jié)構(gòu)。
1.4 ?系統(tǒng)軟件設(shè)計(jì)
系統(tǒng)采用數(shù)據(jù)分析技術(shù)發(fā)現(xiàn)隱藏的信息[6],及時(shí)了解客戶的行為和要求,提高企業(yè)產(chǎn)品銷量。數(shù)據(jù)整理和數(shù)據(jù)清洗這兩個(gè)過程能充分了解客戶的行為和要求。
1.4.1 ?數(shù)據(jù)整理流程
企業(yè)實(shí)施大數(shù)據(jù)挖掘過程中需對(duì)關(guān)聯(lián)信息與冗余信息進(jìn)行整理[7],消除關(guān)聯(lián)信息,具體流程如圖4所示。
1.4.2 ?數(shù)據(jù)清洗
為了降低數(shù)據(jù)收集樣本數(shù)量,排除冗余信息,需對(duì)數(shù)據(jù)進(jìn)行清理。數(shù)據(jù)清洗的過程分為三部分:對(duì)相同字段進(jìn)行清洗;對(duì)意義相同的字段進(jìn)行清洗;對(duì)無價(jià)值的數(shù)據(jù)進(jìn)行清洗[8]。


分析表1可知,本文系統(tǒng)測(cè)試的預(yù)期結(jié)果與實(shí)際結(jié)果是相同的,本文系統(tǒng)滿足企業(yè)需求并且效果顯著。
實(shí)驗(yàn)為測(cè)試本文系統(tǒng)的運(yùn)行性能,分別采用本文系統(tǒng)、基于Web網(wǎng)絡(luò)大數(shù)據(jù)分類的智能企業(yè)管理系統(tǒng)和基于大數(shù)據(jù)集群架構(gòu)的智能企業(yè)管理系統(tǒng)在相同的實(shí)驗(yàn)環(huán)境中進(jìn)行系統(tǒng)吞吐量和系統(tǒng)響應(yīng)時(shí)間的測(cè)試,結(jié)果如圖5和圖6所示。

由圖5可知本文系統(tǒng)最高吞吐量為290 000 Mb/s;其他兩個(gè)系統(tǒng)的最高吞吐量分別為230 000 Mb/s和220 000 Mb/s,所以本文系統(tǒng)的吞吐量高于其他兩個(gè)系統(tǒng)。由圖6可知,本文系統(tǒng)的平均響應(yīng)時(shí)間為3.5 s,其他兩個(gè)系統(tǒng)的平均響應(yīng)時(shí)間分別為8 s和8.5 s,本文系統(tǒng)的平均響應(yīng)時(shí)間比其他系統(tǒng)的平均響應(yīng)時(shí)間快。這些數(shù)據(jù)說明本文系統(tǒng)吞吐量高,平均響應(yīng)時(shí)間快,可實(shí)現(xiàn)高效率的企業(yè)智能管理。
本文以大數(shù)據(jù)為基礎(chǔ)設(shè)計(jì)智能企業(yè)管理系統(tǒng),采用管理應(yīng)用創(chuàng)新模塊,通過多屏互動(dòng)等方式實(shí)現(xiàn)企業(yè)智能管理應(yīng)用創(chuàng)新,通過數(shù)據(jù)整理和數(shù)據(jù)清洗兩種數(shù)據(jù)分析技術(shù)發(fā)現(xiàn)隱藏的信息,及時(shí)了解客戶的行為和要求,提高企業(yè)產(chǎn)品銷量。經(jīng)實(shí)驗(yàn)證明,系統(tǒng)測(cè)試的預(yù)期結(jié)果與實(shí)際結(jié)果相同,使用本文系統(tǒng)進(jìn)行企業(yè)管理時(shí)系統(tǒng)最高吞吐量為290 000 Mb/s,平均響應(yīng)時(shí)間為3.5 s。實(shí)驗(yàn)數(shù)據(jù)表明,使用本文系統(tǒng)能夠?qū)崿F(xiàn)高效的企業(yè)智能管理。
參考文獻(xiàn)
[1] 范敏,劉軍.基于Web網(wǎng)絡(luò)大數(shù)據(jù)分類系統(tǒng)的設(shè)計(jì)與實(shí)現(xiàn)[J].電子設(shè)計(jì)工程,2018,26(8):106?109.
FAN Min, LIU Jun. Design and implementation of large data classification system based on Web network [J]. Electronic design engineering, 2018, 26(8): 106?109.
[2] 徐武沙.基于網(wǎng)絡(luò)環(huán)境下企業(yè)管理系統(tǒng)設(shè)計(jì)[J].現(xiàn)代工業(yè)經(jīng)濟(jì)和信息化,2016,6(21):87?88.
XU Wusha. Design of enterprise management system based on network environment [J]. Modern industrial economy and informationization, 2016, 6(21): 87?88.
[3] 王華濤.基于大數(shù)據(jù)集群架構(gòu)的數(shù)據(jù)管理系統(tǒng)設(shè)計(jì)[J].電腦與信息技術(shù),2016,24(6):29?33.
WANG Huatao. Design of data management system based on data cluster architecture [J]. Computer and information technology, 2016, 24(6): 29?33.
[4] SABITHA M S, VIAYALAKSHMI S, SRE RATHIKAA R M. Big data management system for the harmonization of enterprise model [C]// Proceedings of International Conference on Computing Technologies and Intelligent Data Engineering. Kovilpatti: IEEE, 2016: 1?6.
[5] 孫仕亮,陳俊宇.大數(shù)據(jù)分析的硬件與系統(tǒng)支持綜述[J].小型微型計(jì)算機(jī)系統(tǒng),2017,38(1):1?9.
SUN Shiliang, CHEN Junyu. Review of hardware and system supports for big data analysis [J]. Journal of Chinese computer systems, 2017, 38(1): 1?9.
[6] ZHOU C, CAO Q. Design and implementation of intelligent manufacturing project management system based on bill of material [J]. Cluster computing, 2018(5): 1?9.
[7] 徐森威,徐驥龍,朱琦楨,等.水利施工企業(yè)管理系統(tǒng)設(shè)計(jì)與實(shí)現(xiàn)[J].浙江水利水電學(xué)院學(xué)報(bào),2016,28(6):77?81.
XU Senwei, XU Jilong, ZHU Qizhen, et al. Design and implementation of management system for water conservancy construction enterprises [J]. Journal of Zhejiang Water Conservancy and Hydropower College, 2016, 28(6): 77?81.
[8] FENG Y. Design and implementation of marketing management system for real estate enterprises based on Internet [C]// Proceedings of International Conference on Robots & Intelligent System. Huaian: IEEE, 2017: 94?97.
[9] AOUF R, ALAWNEH A A, ABBOUD H A, et al. Integration of location?based information into mobile learning management system to verify scientific formulas in informal learning environment [J]. International journal of intelligent enterprise, 2017, 4(1): 46?57.
[10] YAO H. Design and implement of food enterprise finance system based on big data [J]. Advance journal of food science and technology, 2016, 10(3): 194?198.