Wednesday, 28 February 2018

“Data is the new plastic” - Data Algorithms in….Plastics Injection Moulding

Hello and welcome back to my “data is the new plastic” blog post series.

In this post I will scrutinize how plastics processing companies, in particular injection moulding operations increase efficiency using algorithms. The main question evolves around which algorithms can we use to improve our processing operations and which data we have to provide to make these operations successful?

“The goal is to turn data into information, and information into insight.” – Carly Fiorina, former executive, president, and chair of Hewlett-Packard Co.

Since decades, quality departments collected data from injection moulding machines for a production run and ensure the defined set of quality levels. Once quality is fulfilled, data will be stored and is available in case of complains. There is much more potential in the data which is stored over time. Algorithms (Wikipedia definition: a process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer”) enable the analysis of large data sets. They allow a descriptive or predictive (future) look at your production facility. Moulders are under constant cost pressure, especially in the Automotive industry. Therefore, applying data analytics might bring you the 2-3% improvement overall needed to keep efficiency innovation alive and competition away. Looking from a broader angle, every task in business operations has a potential digital component. Average business tasks will disappear into the cloud.

Based on the DataIkuBlog post , I made a table to give you an overview on algorithms used in data mining:

Following a list of scientific publication topics on how algorithms can be applied in plastics processing is shown:
1) Taguchi method: optimization of injection moulding machine parameters and quality characteristics [1-3].

2) Artificial Neural Networks: process modelling, parameter optimization and quality production [4-8].

3) Fuzzy logic: predict flash of mixed materials [9].

4) Generic Algorithms in injection moulding: optimize the parameters of the process [10].

5) Response Surface Methodology: create a non-linear model of the process which is used to control process [11-12].

6) Rule based expert systems and case based reason techniques: design plastic injection processes [13-14].

7) Linear Regression Models: predict production [15].

8) Support Vector Machines: quality monitoring of the process [16]

Now a quick reality check: how does it look like in industrial practice?

Most of the algorithms listed before are not yet implemented completely in real industrial environment, however current innovation speed together with the Open Platform Communication (OPC) standard (Euromap 77) is changing this. Data will be more accessible and turned into valuable information used to improve efficiency, predictions and decisions.
A positive outlook for processors, or putting it all together:
Analyzing your business data will lead to a massive productivity advantage and as such more revenue for your business. Extra capital can be used then to expand other fields of the Industrial Internet of Things. Over this route, new digital business models are generated over which you can attract more customers. Platform business models, especially B-2-B platforms (“Platform as a Service”) show here an enormous potential in plastics industry. They need to be open and independent. Currently more and more B-2-C platforms are sweeping over and interest in B-2-B platforms is increasing. At the end, new business models should be explored together by using deep technology (artificial intelligence to just name one). Plastics companies should at least in a theoretical experiment think how a platform could look like. It could look like this:

Example of platform economy in plastics industry

There will be main customers such as OEMs and Tier-1s on one hand and all kind of plastics business such as tooling, moulding and service providers on the other. The latter can function in a complementary manner. For example, when a part from a moulding company is ordered, a simulation and a suggestion for optimal material selection can be provided by the first when possible or by a complementary external party connected to the main platform (comprehensive global scale). In this way, customer service will be stronger than ever. Once again, OPC will be the enabler of such services.  
Data analytics as interface between IT and plastics engineering will enable companies and customers to strive their ventures. On the other hand, neglecting data analytics can be fatal for business. First changes need to be done now.
“If we have data, let’s look at data. If all we have are opinions, let’s go with mine.” – Jim Barksdale, former Netscape CEO
“Without data you're just another person with an opinion.” - W. Edwards Deming
Thanks for reading and till next time!
Herwig Juster
New to my Findoutaboutplastics Blog – check out the start here section.

[1] Chung-Feng J. K., Te-Li S, “Optimization of multiple qualitycharacteristics for polyether ether ketone injection molding process”,Fibers and Polymers, Dec 2006, Volume 7, Issue 4, pp 404-413.
[2] Chung-Feng J. K., Te-Li S., “Optimization of Injection Molding Processing Parameters for LCD Light-Guide Plates”, Journal of Materials Engineering and Performance, Oct 2007, Volume 16, Issue 5, pp 539-548.
[3] Oktem, Tuncay Erzurumlu, Ibrahim Uzman, “Application of Taguchi optimization technique in determining plastic injection molding process parameters for a thin-shell part”, Materials & Design, Volume 28, Issue 4, 2007, Pages 1271–1278.
[4] Jie-Ren S., “Optimization of injection molding process for contour distortions of polypropylene composite components by a radial basis neural network”, The International Journal of Advanced Manufacturing Technology, April 2008, Volume 36, Issue 11-12, pp 1091-1103.
[5] Sadeghi B.H.M., “A BP-neural network predictor model for plastic injection molding process” Journal of Materials Processing Technology, Volume 103, Issue 3, 17 July 2000, Pages 411–416.
[6] Ozcelik B., Erzurumlu T.,” Comparison of the warpage optimization in the plastic injection molding using ANOVA, neural network model and genetic algorithm”, Journal of Materials Processing Technology, Volume 171, Issue 3, 1 February 2006, Pages 437–445.
[7] Shi F., Lou Z.L., Zhang Y.Q., F. Shi, Lu J.G., “Optimisation of Plastic Injection Moulding Process with Soft Computing”, The International Journal of Advanced Manufacturing Technology, June 2003, Volume 21, Issue 9, pp 656-661.
[8] Chen W.C, Tai P.H., Wang M.W, Deng W.J., Chen C.T. “A neural network-based approach for dynamic quality prediction in a plastic injection molding process”. Expert Systems with Applications, Volume 35, Issue 3, October 2008, Pages 843–849.
[9] Zhu J., Chen J.C., “Fuzzy neural network-based in-process mixed material-caused flash prediction (FNN-IPMFP) in injection molding operations” The International Journal of Advanced Manufacturing Technology, June 2006, Volume 29, Issue 3-4, pp 308-316.
[10] Shi F., Lou Z.L., Zhang Y.Q, Lu J.G, “Optimisation of Plastic Injection Moulding Process with Soft Computing” The International Journal of Advanced Manufacturing Technology. June 2003, Volume 21, Issue 9, pp 656-661.
[11] Mathivanan D., Parthasarathy N.S. “Prediction of sink depths using nonlinear modeling of injection molding variables“, The International Journal of Advanced Manufacturing Technology, August 2009, Volume 43, Issue 7-8, pp 654-663.
[12] Mathivanan D., Parthasarathy N.S., “Sink-mark minimization in injection molding through response surface regression modeling and genetic algorithm” The International Journal of Advanced Manufacturing Technology, December 2009, Volume 45, Issue 9-10, pp 867-874.
[13] Kwong C.K., Smith G.F., “A computational system for process design of injection moulding: Combining a blackboard-based expert system and a case-based reasoning approach” The International Journal of Advanced Manufacturing Technology, 1998, Volume 14, Issue 5, pp 350-357.
[14] Shelesh-Nezhad K., Siores E. “An intelligent system for plastic injection molding process design” Journal of Materials Processing Technology, Volume 63, Issues 1–3, January 1997, Pages 458–462.
[15] DasNeogi P., Cudney E., Adekpedjou A., “Comparing the Predictive Ability of T-Method and Cobb-Douglas Production Function for Warranty Data”, ASME 2009 International Mechanical Engineering Congress and Exposition (IMECE2009) , November 13–19, 2009 , Lake Buena Vista, Florida, USA.
[16] Ribeiro B., “Support vector machines for quality monitoring in a plastic injection molding process”, IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Aug 2005, Volume 35, Issue 3, Pages 401 – 410.

No comments:

Post a comment