A Machine Learning Approach to Shipment Consolidation


  • Bas van Andel Maastricht University




Machine Learning, Shipment Consolidation, DHL, cluster analysis, poisson process


This research analyzes the current approach used by a client of DHL LLP for the transportation of shipments from suppliers to production sites. As a result of this analysis, several improvements are proposed that can be used to reduce the costs of transportation in the logistics network. Focus is put on shipment consolidation, rather than on the rerouting of shipments. In this paper, two consolidation methods are introduced. Cluster analysis groups together suppliers that are geographically close and generally ship to the same production sites, and a time-based policy that introduces a maximum waiting time for shipments before they are released. While only marginal improvements are obtained when applying these techniques indepen- dently, a combination of the two provides a powerful synergy. A trade-off between the savings and proportion of on-time shipments arises, when a maximum waiting time is introduced. The potential savings, therefore, depends on the company’s tol- erance for late shipments. While the proposed techniques work in theory, practical and organizational challenges emerge when applying them in the real world. 


annual report. (2017). Deutsche Post DHL Group. Retrieved from http://www.dpdhl.com/content/dam/dpdhl/Investors/Events/Reporting/2017/FY2016/ DPDHL 2016 Annual Report.pdf

Algorithms for k-means clustering. (2013, Jan). University of California San Diego. Retrieved from https://cseweb.ucsd.edu/ dasgupta/291geom/kmeans.pdf

Aloise, D., Deshpande, A., Hansen, P., Popat, P. (2009). Np-hardness of euclidean sum-of-squares clustering. Machine learning, 75(2), 245–248.

Arthur, D., Vassilvitskii, S. (2007). k-means++: The advantages of careful seeding. In Proceedings of the eighteenth annual acm-siam symposium on discrete algorithms (pp. 1027–1035).

Cetinkaya, S., Bookbinder, J. H. (2003). Stochastic models for the dispatch of con- solidated shipments. Transportation Research Part B: Methodological, 37(8), 747–768.

Deng, N. (2013). Shipment consolidation and distribution models in the international supply chain (Unpublished doctoral dissertation). University of Missouri–Columbia.

Desrochers, M., Desrosiers, J., Solomon, M. (1992). A new optimization algorithm for the vehicle routing problem with time windows. Operations research, 40(2), 342–354.

Durrett, R. (2011). Essentials of stochastic processes. Springer.

Dutch, S. (2016, Jan). Converting utm to latitude and longitude. University of Wis- consin - Green Bay. Retrieved from https://www.uwgb.edu/dutchs/UsefulData/ UTMFormulas.HTM

Fisher, R. A. (1936). The use of multiple measurements in taxonomic problems. An- nals of eugenics, 7(2), 179–188.

Foster, P. (n.d.). Maximum likelihood estimation. The University of Manchester. Retrieved from http://www.maths.manchester.ac.uk/peterf/CSI ch4 part1.pdf

Gath, I., Geva, A. B. (1989). Unsupervised optimal fuzzy clustering. IEEE Transac- tions on pattern analysis and machine intelligence, 11(7), 773-780.

Hartigan, J. A., Wong, M. A. (1979). Algorithm as 136: A k-means clustering algorithm. Journal of the Royal Statistical Society. Series C (Applied Statistics), 28(1), 100–108.

Johnson, R. (2014). Clustering and association. University of Notre Dame. Retrieved from http://www3.nd.edu/rjohns15/cse40647.sp14/www/content/lectures/ 13%20-%20Hierarchical%20Clustering.pdf

Kodinariya, T. M., Makwana, P. R. (2013). Review on determining number of cluster in k-means clustering. International Journal, 1(6), 90–95.

Lead logistics partner (LLP). (n.d.). Deutsche Post DHL Group. Retrieved from http://www.dhl.com/en/logistics/lead logistics provider llp.html

Mutlu, F., Cetinkaya, S. i. l., Bookbinder, J. H. (2010). An analytical model for com- puting the optimal time-and-quantity-based policy for consolidated shipments. Iie Trans- actions, 42(5), 367–377.

Slonim, N., Aharoni, E., Crammer, K. (2013). Hartigan’s k-means versus lloyd’s k-means-is it time for a change? In Ijcai.

Tan, P.-N., Steinbach, M., Kumar, V. (2013). Data mining cluster analysis: basic concepts and algorithms. Introduction to data mining.

Unsupervised learning. (n.d.). The University of Iowa. Retrieved from http://homepage .divms.uiowa.edu/ hzhang/c145/notes/chap18b.pdf

Zhao, Q., Hautamaki, V., Franti, P. (2008). Knee point detection in BIC for detecting the number of clusters. In International conference on advanced concepts for intelligent vision systems (pp. 664–673).