Purpose

Modern production facilities produce large amounts of data. The computational framework often referred to as big data analytics has greatly improved the capabilities of analyses of large data sets. Many manufacturing companies can now seize this opportunity to leverage their data to gain competitive advantages for continuous improvement. Six Sigma has been among the most popular approaches for continuous improvement. The data-driven nature of Six Sigma applied in a big data environment can provide competitive advantages. In the traditional Six Sigma implementation – define, measure, analyze, improve and control (DMAIC) problem-solving strategy where a human team defines a project ahead of data collection. This paper aims to propose a new Six Sigma approach that uses massive data generated to identify opportunities for continuous improvement projects in a manufacturing environment in addition to human input in a measure, define, analyze, improve and control (MDAIC) format.

Design/methodology/approach

The proposed Six Sigma strategy called MDAIC starts with data collection and process monitoring in a manufacturing environment using system-wide monitoring that standardizes continuous, attribute and profile data into comparable metrics in terms of “traffic lights.” The classifications into green, yellow and red lights are based on pre-control charts depending on how far a measurement is from its target. The proposed method monitors both process parameters and product quality data throughout a hierarchical production system over time. An attribute control chart is used to monitor system performances. As the proposed method is capable of identifying changed variables with both spatial and temporal spaces, Six Sigma teams can easily pinpoint the areas in need to initiate Six Sigma projects.

Findings

Based on a simulation study, the proposed method is capable of identifying variables that exhibit the biggest deviations from the target in the Measure step of a Six Sigma project. This provides suggestions of the candidates for the improvement section of the proposed MDAIC methodology.

Originality/value

This paper proposes a new approach for the identifications of projects for continuous improvement in a manufacturing environment. The proposed framework aims to monitor the entire production system that integrates all types of production variables and the product quality characteristics.

中文翻译：

---

MDAIC –大数据环境中的六西格码实施策略

目的

现代化的生产设施可产生大量数据。通常称为大数据分析的计算框架极大地提高了对大数据集的分析能力。现在，许多制造公司可以抓住这一机会，利用其数据来获得竞争优势，以进行持续改进。六西格码（Six Sigma）已成为最受欢迎的持续改进方法。在大数据环境中应用六西格码的数据驱动特性可以提供竞争优势。在传统的六西格码实施中–定义，测量，分析，改进和控制（DMAIC）解决问题的策略，其中人工团队在数据收集之前定义项目。

设计/方法/方法

拟议中的称为MDAIC的六西格玛策略开始于在制造环境中使用系统范围的监视对数据进行采集和过程监视，该监视范围将连续数据，属性数据和配置文件数据标准化为“交通信号灯”中的可比指标。绿灯，黄灯和红灯的分类基于预控制图，具体取决于测量距目标的距离。所提出的方法会随着时间的推移监视整个分层生产系统中的过程参数和产品质量数据。属性控制图用于监视系统性能。由于所提出的方法能够识别时空变化的变量，因此六个Sigma团队可以轻松地确定需要启动六个Sigma项目的区域。

发现

在仿真研究的基础上，提出的方法能够识别在6西格玛项目的“测量”步骤中与目标存在最大偏差的变量。这为拟议的MDAIC方法论的改进部分的候选人提供了建议。

创意/价值

本文提出了一种识别项目的新方法，以在制造环境中进行持续改进。提议的框架旨在监视集成了所有类型的生产变量和产品质量特征的整个生产系统。