Abstract
In batch processing, process control is typically carried out comparing trajectories of process variables with those in an in-control set of batches that yielded products within specifications. However, one strong assumption of these schemes is that all batches have equal duration and are synchronized, which is often not satisfied in practice. To overcome that, dynamic time warping (DTW) methods may be used to synchronize stages and align the duration of batches. In this paper, three DTW methods are compared using supervised classification through the k-nearest neighbor technique to determine the in-control set in a milk chocolate conching process. Four variables were monitored over time and a set of 62 batches with durations between 495 and 1170 min was considered; 53% of the batches were known to be conforming based on lab test results and experts’ evaluations. All three DTW methods were able to promote the alignment and synchronization of batches; however, the KMT method (Kassidas et al. in AIChE J 44(4):864–875, 1998) outperformed the others, presenting 93.7% accuracy, 97.2% sensitivity, and 90.3% specificity in batch classification as conforming and non-conforming. The drive current of the main motor was the most consistent variable from batch to batch, being deemed the most important to promote alignment and synchronization of the chocolate conching dataset.
Similar content being viewed by others
References
Afoakwa EO, Paterson A, Fowler M, Vieira J (2008) Relationship between rheological, textural and melting properties of dark chocolate as influenced by particle size distribution and composition. Eur Food Res Technol 227:1215–1223
Alpaydın E (2010) Introduction to machine learning. The MIT Press, Cambridge
Anzanello MJ, Albin SL, Chaovalitwongse WA (2009) Selecting the best variables for classifying production batches into two quality levels. Chemom Intell Lab Syst 97:111–117
Bhatia N, Vandana A (2010) Survey of nearest neighbor techniques. Int J Comput Sci Inf Secur 8(2):302–305
Bolenz S, Thiessenhusen T, Schäpe R (2003) Fast conching for milk chocolate. Eur Food Res Technol 218:62–67
Bolenz S, Kutschke E, Lipp E (2008) Using extra dry milk ingredients for accelerated conching of milk chocolate. Eur Food Res Technol 227:1677–1685
Bolenz S, Manske A, Langer M (2014) Improvement of process parameters and evaluation of milk chocolates made by the new coarse conching process. Eur Food Res Technol 238:863–874
Bühler A (2010) Operating Instructions ELK/DÜC conches. Uzwil, Switzerland
Cidell JL, Alberts HC (2006) Constructing quality: the multinational histories of chocolate. Geoforum 37(6):999–1007
Cover TM, Hart PE (1967) Nearest neighbor pattern classification. IEEE Trans Inf Theory 13(1):21–27
Di Mattia C, Martuscelli M, Sacchetti G, Beheydt B, Mastrocola D, Pittia P (2014) Effect of different conching processes on procyanidin content and antioxidant properties of chocolate. Food Res Int 63:367–372
Dora M, Kumar M, Van Goubergen D, Molnar A, Gellynck X (2013) Food quality management system: reviewing assessment strategies and a feasibility study for European food small and medium-sized enterprises. Food Control 31:607–616
Duda RO, Hart PE, Stork DG (2001) Pattern classification. Wiley, New York
Franke K, Tscheuschner HD (1991) Modelling of the continuous high shear rate conching process for chocolate. J Food Eng 14:103–115
García-Muñoz S, Kourti T, MacGregor JF, Mateos AG, Murphy G (2003) Troubleshooting of an industrial batch process using multivariate methods. Ind Eng Chem Res 42:3592–3601
Glicerina V, Balestra F, Dalla Rosa M, Romani S (2013) Rheological, textural and calorimetric modifications of dark chocolate during process. J Food Eng 119:173–179
Glicerina V, Balestra F, Dalla Rosa M, Romani S (2015) Effect of manufacturing process on the microstructural and rheological properties of milk chocolate. J Food Eng 145:45–50
González-Martínez JM, Ferrer A, Westerhuis JÁ (2011) Real-time synchronization of batch trajectories for on-line multivariate statistical process control using dynamic time warping. Chemom Intell Lab Syst 105:195–206
Jackson JE, Mudholkar GS (1979) Control procedures for residuals associated with principal component analysis. Technometrics 21(3):341–349
Kassidas A, MacGregor JF, Taylor PA (1998) Synchronization of batch trajectories using dynamic time warping. AIChE J 44(4):864–875
Kourti T (2003a) Multivariate dynamic data modeling for analysis and statistical process control of batch processes, start-ups and grade transitions. J Chemom 17:93–109
Kourti T (2003b) Abnormal situation detection, three-way data and projection methods; robust data archiving and modeling for industrial applications. Ann Rev Control 27:131–139
Kourti T, MacGregor JF (1996) Multivariate SPC methods for process and product monitoring. J Qual Technol 28:409–428
Kuang TH, Yan Z, Yao Y (2015) Multivariate fault isolation via variable selection in discriminant analysis. J Process Control 35:30–40
Lim SAH, Antony J, Albliwi S (2014) Statistical process control (SPC) in the food industry—a systematic review and future research agenda. Trends Food Sci Technol 37:137–151
Lu N, Gao F, Yang Y, Wang F (2004) PCA-based modelling and on-line monitoring strategy for uneven-length batch processes. Ind Eng Chem Res 43:3343–3352
MacGregor JF (1997) Using on-line process data to improve quality: challenges for statisticians. Int Stat Rev 65:309–323
Nomikos P, MacGregor JF (1994) Monitoring batch processes using multiway principal component analysis. AlChE J 40(8):1361–1375
Nomikos P, MacGregor JF (1995) Multivariate SPC charts for monitoring batch processes. Technometrics 37(1):41–59
Owusu M, Petersen MA, Heimdal H (2012) Effect of fermentation method, roasting and conching conditions on the aroma volatiles of dark chocolate. J Food Process Preserv 36:446–456
Prawira M, Barringer SA (2009) Effects of conching time and ingredients on preference of milk chocolate. J Food Process Preserv 33:571–589
Ramaker HJ, Van Sprang ENM, Westerhuis JA, Smilde AK (2003) Dynamic time warping of spectroscopic batch data. Anal Chim Acta 498:133–153
Torres-Moreno M, Tarrega A, Costell E, Blanch C (2012) Dark chocolate acceptability: influence of cocoa origin and processing conditions. J Sci Food Agric 92:404–411
Woodall W, Montgomery D (2014) Some current directions in the theory and application of statistical process monitoring. J Qual Technol 46:78–94
Acknowledgements
The authors would like to thank the “Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPQ)” for their financial support, and the chocolate manufacturer that provided the dataset from a conching batch operation. The authors further wish to acknowledge the anonymous referees for their valuable comments that helped to improve this work significantly.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Peres, F.A.P., Peres, T.N., Fogliatto, F.S. et al. Strategies for synchronizing chocolate conching batch process data using dynamic time warping. J Food Sci Technol 57, 122–133 (2020). https://doi.org/10.1007/s13197-019-04037-5
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13197-019-04037-5