(37c) Challenges for Big Data Use of Production Process Data | AIChE

(37c) Challenges for Big Data Use of Production Process Data

Authors 

Sturnfield, J. - Presenter, The Dow Chemical Company
Mendenhall, J. D., The Dow Chemical Company
Trahan, D. W., The Dow Chemical Company
D'Ottaviano, F., The Dow Chemical Company
Shuang, B., Dow
There are a number of initiatives for the future of manufacturing such as Industry 4.0(1), Digital Transformation Initiative(2), and Smart Manufacturing Leadership Coalition(3). These bring together a variety of concepts to create a seamless connection between the management of various data sources, advanced computer simulations/analysis, and adaptive user/machine/organization interactions. All of these include process data as part of their corporate integration, but there are a number of issues with this type of data that will need to be address if it is going to be made transparent to the user. In particular, this is the incorporation of knowledge within Big Data(4). This presentation will look at some of these issues and consider examples where these issues impact the understanding of the data.

In particular, the presentation will consider issues of the compression of production data, examples about various types of errors in the measurements, the time delays within a process, and the high correlation of process operation. The knowledge applied to these issues includes the details about the processes that ranges from the physical information about the equipment to the materials and conditions used to the theoretical understanding of the system. The knowledge needed also includes the type of sensors that are used for the measurements and the management and sampling of the data.

In addition, there is a need to understand the various intended use of the data and the appropriate analysis for that intention. The computations should be different if the intent is to predict future action compared to an intent of understanding why a certain defect occurred. This is a difference that is often ignored when analyzing data.

The examples are derived from actual production data to provide actual insight to how these issues have been traditional handled and provide some examples of how advanced analytics has helped with these issues. The intention is to encourage further development of working with production data that deals with these issues and that could be truly fulfill some of the vision of data transparency and advance analytics for manufacturing’s next evolution.

  1. Kagermann H, Wahlster W, Helbig J. Recommendations for implementing the strategic initiative INDUSTRIE 4.0. Final report of the Industrie 4.0 Working Group. Acatech- National Academy of Science and Engineering; 2013.
  2. World_Economic_Forum. Digital Transformation Initiative Chemistry and Advanced Materials Industry. Geneva, Switzerland: World Economic Forum; 2017.
  3. Davis J, Edgar T, Porter J, Bernaden J, Sarli M. Smart manufacturing, manufacturing intelligence and demand-dynamic performance. Computers & Chemical Engineering. 2012;47(Supplement C):145-56.
  4. Wuest T, Weimer D, Irgens C, Thoben K-D. Machine learning in manufacturing: advantages, challenges, and applications. Production & Manufacturing Research. 2016;4(1):23-45.