基于PLS-Aenet的多工序制造过程关键质量特性识别

doi:10.16381/j.cnki.issn1003-207x.2021.1658

Abstract

Abstract:

The complex multistage manufacturing process contains many quality characteristics which are characterized by high dimensional multicollinearity and group effect. In the actual production process， subjected to the cost and technology constraints， all quality characteristics cannot be monitored. How to identify the key variables from a lot of quality characteristics with high dimensional multicollinearity and group effect has become increasingly necessary. To solve this problem， The Partial Least Squares- Adaptive Elastic Net （PLS-Aenet） method is applied to model and analyze the complex multistage manufacturing process and identify the key quality characteristics of the process. The PLS-Aenet method is used and the state space idea is integrated to model and identify the key quality characteristics of complex multistage manufacturing processes， and the specific steps are given to identify the key quality characteristics based on the PLS-Aenet Net method. Finally， the applicability and effectiveness of the PLS-Aenet Net method and the variable selection method such as Ridge Regression、Lasso、PLS-Lasso、Elastic Net method in the identification of key quality characteristics of the complex multistage manufacturing process are compared through simulation analysis. The simulation and real case results show that the PLS-Aenet method achieves a more effective performance when the key quality characteristics are strongly correlated with each other.

Key words: multistage manufacturing process, complex product, state space model, PLS-Aenet, key quality characteristics

CLC Number:

F273

Ning Wang,Shuke Tian,Yumin Liu,Zheyun Zhao. Identification of Key Quality Characteristics in Multistage Manufacturing Process Based on PLS-Aenet[J]. Chinese Journal of Management Science, 2024, 32(4): 271-278.

Figures/Tables 8

过程质量特性	真值	$ρ = 0.3$					$ρ = 0.6$					$ρ = 0.9$
过程质量特性	真值	岭回归	LASSO	弹性网	PLS-LASSO	PLS-Aenet	岭回归	LASSO	弹性网	PLS-LASSO	PLS-Aenet	岭回归	LASSO	弹性网	PLS-LASSO	PLS-Aenet
u₁	20	15.885	16.991	19.217	22.510	22.506	17.598	19.898	17.521	22.929	22.526	19.153	20.438	19.374	21.252	23.017
u₂	20	14.754	18.866	18.227	23.457	25.452	17.912	20.640	19.044	22.910	21.676	17.959	21.171	20.826	22.039	22.335
u₃	1	0.099	0.000	0.000	0.000	0.000	1.501	0.101	0.000	0.000	0.000	4.223	0.000	0.467	0.000	0.000
u₄	1	1.039	1.671	0.000	0.000	0.000	1.584	0.000	0.000	0.000	0.717	4.063	0.000	0.414	0.000	0.000
u₅	20	15.301	17.437	18.260	22.810	23.102	17.636	19.262	19.442	22.638	22.506	15.650	20.003	19.376	19.522	19.516
u₆	20	15.970	17.296	17.645	21.951	22.612	18.334	19.349	18.434	23.781	21.807	16.329	15.637	16.052	23.332	23.182
u₁₂	10	10.747	10.287	10.387	10.565	9.955	10.357	13.100	12.236	10.722	11.635	6.799	24.401	9.621	0.000	15.813
u₁₃	1	1.162	1.891	1.581	0.000	0.854	1.707	1.947	1.881	0.000	0.000	2.981	3.454	3.396	0.000	0.000
u₁₄	1	1.536	0.773	1.174	0.667	0.634	0.463	1.121	1.643	0.000	0.632	1.920	3.728	1.258	0.000	0.000
u₁₅	1	2.439	1.777	1.204	0.000	0.000	2.020	0.342	1.331	0.000	0.000	3.259	0.000	2.141	0.000	0.000
u₁₆	1	2.031	1.578	1.352	0.000	0.000	1.638	1.060	1.585	0.000	0.000	3.437	0.000	2.043	0.000	0.000
u₂₃	1	2.631	1.278	2.466	0.017	0.000	2.344	1.584	1.790	0.000	0.281	2.895	0.534	2.511	0.000	0.000
u₂₄	1	1.798	0.939	1.368	0.036	0.260	1.304	1.884	2.139	0.000	0.000	2.029	1.229	1.755	0.000	0.000
u₂₅	1	2.721	0.774	1.317	0.000	0.000	1.979	2.352	0.007	0.000	0.000	3.512	5.257	3.499	0.000	0.000
u₂₆	1	2.121	1.876	1.341	0.000	0.000	1.423	0.079	1.711	0.002	0.000	3.808	3.209	3.331	0.000	0.000
u₃₄	1	0.229	1.437	1.286	2.943	3.199	0.154	0.814	0.354	0.073	0.000	0.067	0.117	0.000	0.000	0.000
u₃₅	1	1.412	1.326	0.945	0.225	0.000	1.478	0.301	2.848	0.641	1.253	1.940	0.0023	0.914	0.000	0.000
u₃₆	1	1.975	0.887	1.289	1.722	0.880	1.089	2.246	0.583	0.000	0.704	2.402	2.933	1.548	0.000	0.000
u₄₅	1	1.567	1.477	2.095	0.418	0.961	3.112	1.029	1.557	0.000	0.000	3.073	1.438	2.346	0.000	0.000
u₄₆	1	1.676	0.903	1.043	0.000	0.776	2.083	0.148	1.986	0.000	0.000	3.761	0.083	3.521	0.000	0.000
u₅₆	10	10.941	10.871	10.999	10.087	9.972	10.844	12.851	11.822	11.501	10.369	7.092	27.672	9.747	0.000	10.438
u₁₁	10	9.490	9.681	9.084	10.077	9.924	9.587	7.552	8.093	10.557	9.282	7.394	0.000	5.929	11.953	8.479
u₂₂	10	8.474	9.435	8.825	10.383	9.546	8.498	7.205	7.983	10.515	11.151	6.153	0.000	5.420	9.110	10.865
u₃₃	1	0.547	0.086	0.076	0.643	0.484	0.524	0.000	0.000	0.000	0.000	0.038	0.000	0.000	0.000	0.000
u₄₄	1	0.464	0.045	0.187	1.083	0.477	0.114	0.000	0.000	0.017	0.000	0.223	0.000	0.000	0.000	0.000
u₅₅	10	9.304	9.465	8.829	10.363	9.932	7.766	6.562	7.785	10.107	10.898	5.726	0.000	6.673	11.499	16.701
u₆₆	10	8.957	9.597	9.430	10.059	9.812	8.881	8.416	8.321	9.516	10.272	7.184	0.007	6.982	15.553	9.513

方法	$ρ = 0.3$			$ρ = 0.6$			$ρ = 0.9$
方法	保留变量数	剔除实际非关键变量数	保留实际关键变量数	保留变量数	剔除实际非关键变量数	保留实际关键变量数	保留变量数	剔除实际非关键变量数	保留实际关键变量数
岭回归	27	0	10	27	0	10	27	0	10
Lasso回归	26	1	10	24	3	10	18	9	6
弹性网回归	25	2	10	23	4	10	24	3	10
PLS-Lasso	19	8	10	14	13	10	9	18	7
PLS-Aenet	19	8	10	15	12	10	10	17	10

References 31

1	方志耕，王欢，董文杰，等.基于可靠性基因库的民用飞机故障智能诊断网络框架设计［J］. 中国管理科学，2018，26（11）：124-131.
	Fang Z G， Wang H， Dong W J， et al. Frame-work design of intelligent fault diagnosis netw-ork for civil aircraft based on reliability gene-bank［J］. Chinese Journal of Management Science，2018，26（11）：124-131.
2	宋明顺，杨铭，方兴华.基于最大熵分布的控制图改进与评价研究［J］. 中国管理科学，2019，27（12）：208-216.
	Song M S， Yang M， Fang X H. Study on control chart improvement and evaluation based on maximum entropy distribution［J］. Chinese Journal of Management Science，2019，27（12）：208-216.
3	王宁，张帅，刘玉敏，等.多工序串并联制造过程关键质量特性识别［J］. 系统工程学报，2019，34（6）：855-866.
	Wang N， Zhang S， Liu Y M， et al. Identification of key quality characteristics in multi-process series parallel manufacturing process［J］. Journal of Systems Engineering，2019，34（6）：855-866.
4	何桢，赵玉忠.全面质量管理中的关键影响因素分析［J］. 统计与决策，2008，23（12）：164-166.
	He Z， Zhao Y Z. Analysis of key influencing factors in total quality management［J］. Statistics and Decision，2008，23（12）：164-166.
5	Chen Y， Fung R Y， Tang J. Rating technical attributes in fuzzy QFD by integrating fuzzy weighted average method and fuzzy expected value operator［J］. European Journal of Operational Research， 2006， 174（3）： 1553-1566.
6	李丽，金嘉琦，姜兴宇，等. 废旧机床再设计关键质量特性的识别方法［J］. 机械设计与制造，2018，55（5）： 151-154.
	Li L， Jin J Q， Jiang X Y， et al. Identification method of key quality characteristics of waste machine tool redesign［J］. Mechanical Design and Manufacturing， 2018， 55（5）： 151-154.
7	Shen H， Wan S. Controlled sequential factorial design for simulation factor screening ［J］. European Journal of Operational Research， 2008， 198（2）： 511-519.
8	汪四水. 基于交叉谱分析法的因子筛选［J］. 数学的实践与认识，2005， 35（11）： 45-51.
	Wang S S. Factor screening based on Cross-Spectrum analysis［J］. Mathematics in Practice and Cognition， 2005， 35（11）： 45-51.
9	Jin N， Zhou S. Data-driven variation source identification for manufacturing process using the eigenspace comparison method［J］. Naval Research Logistics （NRL）， 2006， 53（5）： 383-396.
10	Wang H， Fang Z G， Wang D A， et al. An integrated fuzzy QFD and grey decision-making approach for supply chain collaborative quality design of large complex products［J］. Computers & Industrial Engineering， 2020，43（140）：106-212.
11	胡桔州，侯木舟，欧阳资生. 数据挖掘的统计方法及其软件实现［J］. 统计与决策，2007，22（5）： 1334-135.
	Hu J Z， Hou M Z， Ouyang Z S. Statistical methods of data mining and its software imp-lementation［J］. Statistics and Decision，2007，22（5）： 1334-135.
12	王宁，闫娜，徐友真，等.复杂多工序制造过程关键质量特性识别［J］.统计与决策，2021，37（8）：177-180.
	Wang N， Yan N， Xu Y Z， et al. Identification of key quality characteristics in complex multi-process manufacturing process［J］.Statistics and Decision Making，2021，37（8）：177-180.
13	徐友真.基于弹性网的多工序制造过程Profile监控研究［D］.郑州：郑州大学，2020.
	Xu Y Z. Research on profile monitoring of multi-process manufacturing process based on elastic net ［D］.Zhengzhou： Zhengzhou University， 2020.
14	闫伟，何祯，田文萌，等.基于IG的复杂产品关键质量特性识别［J］.工业工程与管理，2012， 17（1）： 70-74.
	Yan W， He Z， Tian W M， et al. Identification of key quality characteristics of complex products based on IG［J］.Industrial Engineering and Management，2012， 17（1）： 70-74.
15	闫伟，何祯，田文萌，等.基于EM的不平衡数据关键质量特性识别［J］.工业工程与管理，2012，17（4）：38-42.
	Yan W， He Z， Tian W M， et al. Identification of key quality characteristics of unbalanced data based on EM［J］.Industrial Engineering and Management，2012，17（4）： 38-42.
16	闫伟，何祯，李岸达.基于CEM-IG算法的复杂产品关键质量特性识别［J］.系统工程理论实践，2014，34（5）： 1230-1236.
	Yan W， He Z， Li A D. Identification of key mass characteristics of complex products based on CEM-IG algorithm［J］.Systems Engineering-Theory & Practice，2014，34（5）： 1230-1236.
17	王化强，牛占文.基于LASSO的复杂产品关键质量特性识别［J］.系统工程，2014，32（6）：137-141.
	Wang H Q， Niu Z W. Identification of key quality characteristics of complex products based on LASSO［J］.Systems Engineering， 2014，32（6）：137-141.
18	Wang H Q， Liang L J， Niu Z W， et al. Identification of CTQs for complex products based on mutual information and improved gravitational search algorithm［J］. Mathematical Problems in Engineering，2015，765985.
19	马丽莎，茅健. 基于熵权法和马氏田口法的关键质量特性识别研究［J］. 轻工机械，2017，34（8）：101-105.
	Ma L S， Mao J. Research on identification of key quality characteristics based on entropy weight method and martens taguchi method［J］. Light Industry Machinery， 2017， 34（8）： 101-105.
20	李岸达，何桢，王庆.基于多目标鲸鱼优化的关键质量特性识别方法［J］.系统工程， 2019， 37（1）：138-146.
	Li A D， He Z， Wang Q. Identification method of key quality characteristics based on Multi-objective Whale Optimization［J］.Systems Engineering， 2019， 37（1）：138-146.
21	王宁，张帅，刘玉敏.基于改进Adaptive Lasso的多工序制造过程关键质量特性识别［J］.运筹与管理，2020，29（6）：210-219.
	Wang N， Zhang S， Liu Y M. Identification of key quality characteristics of multi-process manufacturing process based on improved Adaptive Lasso［J］.Operations Research and Management，2020，29（6）：210-219.
22	梁角虎.多参数耦合制造过程质量控制方法的研究［D］.兰州：兰州理工大学，2011.
	Liang J H. Research on quality control method of multi-parameter coupled manufacturing process［D］. Lanzhou： Lanzhou University of Technology， 2011.
23	Jin J， Shi J. State space modeling of sheet metal assembly for dimensional control［J］. ASME Transactions， Journal of Manufacturing Science and Engineering， 1999，121（4）：756-762.
24	Hoerl A， Kennard R. Ridge regression： biased estimation for nonorthogonal problems［J］. Technometrics， 1970，12（1）：55-67.
25	宋瑞琪，朱永忠，王新军.高维数据中变量选择研究［J］.统计与决策，2019，35（2）：13-16.
	Song R Q， Zhu Y Z， Wang X J. Variable selection in high-dimensional data［J］.Statistics and Decision Making，2019，35（2）：13-16.
26	Tibshirain R. Regression shrinkage and selection via the lasso［J］. Journal of the Royal Statistical Society Series B： Statistical Methodology， 1996， 58（1）： 267-288.
27	李根，邹国华，张新雨.高维模型选择方法综述［J］.数理统计与管理，2012，31（4）：640-658.
	Li G， Zou G H， Zhang X Y. A review of high-dimensional model selection methods［J］.Math-ematical Statistics and Management，2012，31（4）：640-658.
28	Zou H， Hastie T. Regression shrinkage and selection via the elastic net with application to microarrays［J］. Journal of the Royal Statistical Society Series B： Statistical Methodology， 2003， 67（2）：301-320.
29	Zou H， Hastie T. Regularization and variable selection via the elastic net［J］. Journal of the Royal Statistical Society Series B：Statistical Methodology， 2005，67（2）： 301-320.
30	王小燕，谢邦昌，马双鸽，等. 高维数据下群组变量选择的惩罚方法综述［J］. 数理统计与管理，2015，34（6）： 978-988.
	Wang X Y， Xie B C， Ma S G， et al. Review of punitive methods for group variable selection in High-Dimensional data［J］. Mathematical Statistics and Management，2015，34（6）： 978-988.
31	Zou H， Zhang H H. On the adaptive elastic-net with a diverging number of parameters［J］. Annals of Statistics， 2009， 37（4）：1733-1751.

变量	u₁₂	u₅₆	u₁₁	u₂₂	u₅₅	u₆₆
u₁₂	1.0000	0.5980	0.9590	0.9626	0.6104	0.5419
u₅₆	0.5980	1.0000	0.5429	0.6089	0.9648	0.9594
u₁₁	0.9590	0.5429	1.0000	0.8543	0.5533	0.4940
u₂₂	0.9626	0.6089	0.8543	1.0000	0.6222	0.5506
u₅₅	0.6104	0.9648	0.5533	0.6222	1.0000	0.8601
u₆₆	0.5419	0.9594	0.4940	0.5506	0.8601	1.0000

质量特性	均值	方差	单位	变量
铅锅温度	517	2.08	℃	u₁
铅勺温度	522	1.8	℃	u₂
动模温度	171	1.66	℃	u₃
定模温度	183	1.96	℃	u₄
大片板栅重量	20.5	0.35	g	u₅
小片板栅重量	0.8	0.4	g	u₆
铅粉重量	975	2	Kg	u₇
酸重量	84	0.2	Kg	u₈
纯水重量	118	1	Kg	u₉
出膏温度	47	3.78	℃	u₁₀
视比重	4.455	0.0029	g/cm³	u₁₁
单片极板重量	352.64	15.14	g	Y

质量特性	弹性网	PLS-Lasso	PLS-Aenet
铅锅温度	2.8967	3.2028	-1.0967
铅勺温度	3.2871	2.7802	3.1282
动模温度	3.2901	4.7131	3.8719
定模温度	0.6712	0.3740	4.7318
大片板栅重量	1.2870	1.4502	1.3827
小片板栅重量	1.3968	-0.9718	1.4576
铅粉重量	-0.1284	0.7116	0.0000
酸重量	-0.8052	0.0000	0.0000
纯水重量	-0.2858	0.0000	0.0000
出膏温度	171.6653	158.7827	-0.4196
视比重	-1.2962	1.3696	177.7787

[1]	Naiding Yang,Yu Wang,Yan Wang,Yanlu Zhang. Evolutionary Game on Technological Innovation of Enterprises in Complex Product R&D Network [J]. Chinese Journal of Management Science, 2024, 32(2): 242-253.
[2]	CHEN Hong-zhuan, HUANG Xin, WANG Wei-ming. Stackelberg Optimization for Spare Parts’s Collaborative Innovation of Complex Products Considering Cost Sharing [J]. Chinese Journal of Management Science, 2022, 30(7): 69-76.
[3]	LI Rui-meng, YANG Nai-ding, LIU Hui, ZHANG Yan-lu. Design Change Risk Propagation for Complex Product Development Projects Considering Organizational Failure and Cooperation [J]. Chinese Journal of Management Science, 2022, 30(10): 265-276.
[4]	ZHANG Qiao-ke, CHEN Hong-zhuan. Incentive Strategy for Complex Product Quality Control Based on Quality Awareness [J]. Chinese Journal of Management Science, 2021, 29(6): 105-114.
[5]	SHEN Genxiang, ZHANG Jingze. Dynamic Nelson-Siegel Term Structure Model with GARCH Error Terms and It’s Applications [J]. Chinese Journal of Management Science, 2021, 29(10): 1-11.
[6]	HE Di, ZHOU Yong. Modeling and Analyzing Liquidity in Stock Market Using Macroeconomic Factors Based on State Space Model [J]. Chinese Journal of Management Science, 2019, 27(5): 42-49.
[7]	CHEN Zhan-duo. Evolutionary of Uncooperative Behaviors Caused by Price Change and Cost Change in Project of Complex Products of System [J]. Chinese Journal of Management Science, 2018, 26(5): 157-168.
[8]	SHEN Gen-xiang, CHEN Ying-zhou. Nelson-Siegel Dynamic Term Structure Model with Double Slope Factors and Its Applications [J]. Chinese Journal of Management Science, 2015, 23(10): 1-10.
[9]	YANG Jin. An Empirical Research of Performance Evaluation Model of Large and Complex Products Manufacturing Cluster Supply Chain [J]. Chinese Journal of Management Science, 2013, 21(6): 64-71.
[10]	ZHU Jie, CHEN Lang-nan. The Performance Evaluation of Dynamic Mutual Funds Based on SSM [J]. Chinese Journal of Management Science, 2013, 21(6): 11-21.
[11]	SONG Yan-qiu, JIA Chuan-liang, GAO Tian-hui. Study on the Effectiveness of Contract Model on Cooperative Innovation in Complex Product System [J]. Chinese Journal of Management Science, 2011, 19(2): 155-160.
[12]	YANG Jin. Investigation on Influencing Mechanism of Supply Chain Management to Evolution of Large-Scale Complex Products Manufacturing Clusters [J]. Chinese Journal of Management Science, 2009, 17(4): 75-83.

Identification of Key Quality Characteristics in Multistage Manufacturing Process Based on PLS-Aenet

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 31

Related Articles 12

Metrics

Comments

Recommended 0