Description
IS420PUAAH1A Reliable Turbine Control Solutions
3 Case Studies on Reducing Scrap Rates
Any product assembled or produced in a factory goes through a series of quality tests to determine whether it needs to be scrapped.
High scrap rates are caused by the opportunity cost of not delivering products to customers in a timely manner, wasted personnel time, wasted
non-reusable parts, and equipment overhead expenses. Reducing scrap rates is one of the main issues manufacturers need to address. Ways to
reduce scrap include identifying the root causes of low product quality.
3.1 Data processing
Root cause analysis begins by integrating all available data on the production line. Assembly lines, workstations, and machines make up the industrial
production unit and can be considered equivalent to IoT sensor networks. During the manufacturing process, information about process status,
machine status, tools and components is constantly transferred and stored. The volume, scale, and frequency of factory production considered in
this case study necessitated the use of a big data tool stack similar to the one shown in Figure 2 for streaming, storing, preprocessing, and
connecting data. This data pipeline helps build machine learning models on batch historical data and streaming real-time data. While batch
data analytics helps identify issues in the manufacturing process, streaming data analytics gives factory engineers regular access to the latest
issues and their root causes. Use Kafka (https://kafka.apache.org) and Spark streaming (http://spark.apache.org/streaming) to transmit real-time
data from different data sources; use Hadoo (http://hadoop.apache.org ) and HBase (https://hbase.apache.org) to store data efficiently; use
Spark (http://spark.apache.org) and MapReduce framework to analyze data. The two main reasons to use these tools are their availability as open
source products, and their large and active developer network through which these tools are constantly updated.
Reviews
There are no reviews yet.