Research Article
BibTex RIS Cite

Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry

Year 2022, Volume: 34 Issue: 3, 415 - 425, 30.09.2022
https://doi.org/10.7240/jeps.1127844

Abstract

Accurate demand forecasting is crucially important to reduce inventory and backlogging cost. In this study, we analyze how
promos, holiday statements, price changes, stock availability and date-time features (weekdays, months etc.) affect the
demand by using several forecasting methods. Data sets were collected for the products of the global pharmaceutical
company providing services in Turkey. Actual daily sales data for 2016, 2017 and 2018 were used in the construction of this
data set. In order to predict the next periods demand, we used four different models which are Holt Winters, Ridge
Regression, Random Forest and Xgboost. We also ensemble those models to improve forecasting accuracy. Next, by
weighting inversely proportional to the error rates of the models, binary, triple and quadruple combinations of the single
models were compared with themselves and the single models. Our numerical results show that the lowest forecasting error
rate was obtained in ensemble models. Particularly, the lowest error rate in individual models was obtained in Random Forest
with 15.7% RMSPE (Root Mean Square Percentage Error) value, and the lowest error rate was obtained with 10.7% RMSPE
value in Holt Winters & Xgboost models combination. Results show that ensemble of several models can increase the
forecasting accuracy. 

References

  • Al-Hafid, M. S., & Hussein Al-maamary, G. (2012). Short term electrical load forecasting using holt-winters method. Al-Rafidain Engineering Journal (AREJ), 20(6), 15–22.
  • Al-Hassan, Y. M. M., & Al-Kassab, M. M. (2000). A comparison between ridge and principal components regression methods using simulation technique. Al Al-Bayt University.
  • Ali, Ö. G., Sayin, S., Van Woensel, T., & Fransoo, J. (2009). SKU demand forecasting in the presence of promotions. Expert Systems with Applications, 36(10), 12340–12348.
  • Biau, G. (2012). Analysis of a random forests model. The Journal of Machine Learning Research, 13(1), 1063–1095. Boulesteix, A.-L., Janitza, S., Kruppa, J., & König, I. R. (2012). Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2(6), 493–507.
  • Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. Chen, C., Liaw, A., Breiman, L., & others. (2004). Using random forest to learn imbalanced data. University of California, Berkeley, 110(1–12), 24.
  • Chen, T. (2014). Introduction to boosted trees. University of Washington Computer Science, 22(115), 14–40.
  • Cook, A. G. (2016). Forecasting for the pharmaceutical industry: models for new product and in-market forecasting and how to use them. Gower.
  • Ferreira, K. J., Lee, B. H. A., & Simchi-Levi, D. (2016). Analytics for an online retailer: Demand forecasting and price optimization. Manufacturing \& Service Operations Management, 18(1), 69–88.
  • Ganesh, S. S., Arulmozhivarman, P., & Tatavarti, R. (2019). Forecasting air quality index using an ensemble of artificial neural networks and regression models. Journal of Intelligent Systems, 28(5), 893–903.
  • Ghousi, R., Mehrani, S., Momeni, M., & Anjomshoaa, S. (2012). Application of data mining techniques in drug consumption forecasting to help pharmaceutical industry production planning. Proceedings of the 2012 International Conference on Industrial Engineering and Operations Management, 3–6.
  • Goodwin, P. (2005). Providing support for decisions based on time series information under conditions of asymmetric loss. European Journal of Operational Research, 163(2), 388–402. https://doi.org/https://doi.org/10.1016/j.ejor.2003.10.039
  • Hand David, J., & Heikki, M. (2001). Principles of data mining (adaptive computation and machine learning). Bradford Books.
  • Hasin, M. A. A., Ghosh, S., & Shareef, M. A. (2011). An ANN approach to demand forecasting in retail trade in Bangladesh. International Journal of Trade, Economics and Finance, 2(2), 154.
  • KARADAVUT, U., Aşır, G.E.N.Ç., TOZLUCA, A., KINACI, İ., AKSOYAK, Ş., PALTA, Ç. and PEKGÖR, A (2005). Nohut Cicer arietinum L. Bitkisinde Verime Etki Eden Bazı Karakterlerin Alternatif Regresyon Yöntemleriyle Karşılaştırılması. Journal of Agricultural Sciences, 11(03), 328–333.
  • Lakshmi Anusha, S., Alok, S., & Shaik, A. (2014). Demand Forecasting for the Indian Pharmaceutical Retail: A Case Study. Journal of Supply Chain Management Systems, 3(2).
  • Liu, L.-M., Bhattacharyya, S., Sclove, S. L., Chen, R., & Lattyak, W. J. (2001). Data mining on time series: an illustration using fast-food restaurant franchise data. Computational Statistics \& Data Analysis, 37(4), 455–476.
  • Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (2008). Forecasting methods and applications. John wiley \& sons.
  • Martin, L.-C., & others. (2019). Machine learning vs. traditional forecasting methods An application to South African GDP.
  • Merkuryeva, G., Valberga, A., & Smirnov, A. (2019). Demand forecasting in pharmaceutical supply chains: A case study. Procedia Computer Science, 149, 3–10.
  • Micajkova, V., Georgieva Svrtinov, V., Petkovski, V., & Esmerova, E. (2018). Selecting appropriate forecast method on the basic of forecast accuracy-pharmaceutical company case study.
  • Moskowitz, H. (1972). The Value of Information in Aggregate Production Planning—A Behavioral Experiment. A I I E Transactions, 4(4), 290–297. https://doi.org/10.1080/05695557208974865
  • Padhy, N., Mishra, D., Panigrahi, R., & others. (2012). The survey of data mining applications and feature scope. ArXiv Preprint ArXiv:1211.5723.
  • Pavlyshenko, B. M. (2019). Machine-learning models for sales time series forecasting. Data, 4(1), 15. Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012).
  • An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93–104.
  • Smits, R. E. H. M., & Boon, W. P. C. (2008). The role of users in innovation in the pharmaceutical industry. Drug Discovery Today, 13(7–8), 353–359.
  • Tosun, T. (2006). Veri madenciliği teknikleriyle kredi kartlarında müşteri kaybetme analizi Fen Bilimleri Enstitüsü.
  • Tugay, R., & Oguducu, S. G. (2020). Demand Prediction Using Machine Learning Methods and Stacked Generalization. arXiv. https://doi.org/10.48550/ARXIV.2009.09756
  • Yıldırım, N. (2010). En küçük kareler, Ridge regresyon ve Robust regresyon yöntemlerinde analiz sonuçlarına aykırı değerlerin etkilerinin belirlenmesi.. Çukurova Üniversitesi Fen Bilimleri Enstitüsü, Yüksek Lisans Tez, Adana.

İlaç Sektöründe Zaman Serisi ve Regresyon Birleşik Modeller ile Talep Tahmini Uygulaması

Year 2022, Volume: 34 Issue: 3, 415 - 425, 30.09.2022
https://doi.org/10.7240/jeps.1127844

Abstract

Doğru talep tahmini, karşılanmayan talep ve stok miktarını azaltmak için büyük önem taşımaktadır. Bu çalışma, yapılan
promosyonların, yıl içi tatil günlerinin, ürünün fiyatında olan değişikliklerin, ürünün stokta bulunup bulunmamasının ve bazı
tarih özelliklerinin (haftanın günleri, aylar, yıllar vb.) birden çok tahmin modelinde kullanılarak talebi nasıl etkilediğinin
analiz edilmesini amaçlamaktadır. Çalışma için, Türkiye'de uzun yıllar hizmet veren bir global ilaç şirketine ait bir ürün
incelenmiştir. Veri seti için 2016, 2017 ve 2018 yıllarına ait günlük satış verileri kullanılmıştır. Gelecek dönemlerin talebini
tahmin etmek için; Holt Winters, Ridge Regression, Rastgele Orman ve Xgboost olmak üzere dört ayrı model kullanılmıştır.
Ayrıca tahmin doğruluğunu arttırmak için dört modelin birbiriyle olan kombinasyonlarından oluşan modeller de
kullanılmıştır. Sonrasında, modellerin hata oranları ile ters orantılı şekilde ağırlıklandırma yapılarak, tekli modellerin ikili,
üçlü ve dörtlü kombinasyonları elde edilmiş ve hata oranları hem kendi aralarında hem de tekli modellerle kıyaslanmıştır.
Sonuçlar, en düşük tahminleme hatalarının birleştirilmiş modellerden elde edildiğini göstermiştir.Oluşturulan tüm modeller
hata oranı bakımından kıyaslandığında, hata oranı en düşük modelimiz %10,7 RMSPE (Kök ortalama Kare Yüzde Hata)
değeri ile Holt Winters ve Xgboost modellerinin kombinasyonlarından oluşan kombinasyon olmuştur. Sonuçlar, birden çok
modelin birlikte kullanılarak talep tahmininin doğruluk oranının artırılabileceğini göstermiştir.

References

  • Al-Hafid, M. S., & Hussein Al-maamary, G. (2012). Short term electrical load forecasting using holt-winters method. Al-Rafidain Engineering Journal (AREJ), 20(6), 15–22.
  • Al-Hassan, Y. M. M., & Al-Kassab, M. M. (2000). A comparison between ridge and principal components regression methods using simulation technique. Al Al-Bayt University.
  • Ali, Ö. G., Sayin, S., Van Woensel, T., & Fransoo, J. (2009). SKU demand forecasting in the presence of promotions. Expert Systems with Applications, 36(10), 12340–12348.
  • Biau, G. (2012). Analysis of a random forests model. The Journal of Machine Learning Research, 13(1), 1063–1095. Boulesteix, A.-L., Janitza, S., Kruppa, J., & König, I. R. (2012). Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2(6), 493–507.
  • Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. Chen, C., Liaw, A., Breiman, L., & others. (2004). Using random forest to learn imbalanced data. University of California, Berkeley, 110(1–12), 24.
  • Chen, T. (2014). Introduction to boosted trees. University of Washington Computer Science, 22(115), 14–40.
  • Cook, A. G. (2016). Forecasting for the pharmaceutical industry: models for new product and in-market forecasting and how to use them. Gower.
  • Ferreira, K. J., Lee, B. H. A., & Simchi-Levi, D. (2016). Analytics for an online retailer: Demand forecasting and price optimization. Manufacturing \& Service Operations Management, 18(1), 69–88.
  • Ganesh, S. S., Arulmozhivarman, P., & Tatavarti, R. (2019). Forecasting air quality index using an ensemble of artificial neural networks and regression models. Journal of Intelligent Systems, 28(5), 893–903.
  • Ghousi, R., Mehrani, S., Momeni, M., & Anjomshoaa, S. (2012). Application of data mining techniques in drug consumption forecasting to help pharmaceutical industry production planning. Proceedings of the 2012 International Conference on Industrial Engineering and Operations Management, 3–6.
  • Goodwin, P. (2005). Providing support for decisions based on time series information under conditions of asymmetric loss. European Journal of Operational Research, 163(2), 388–402. https://doi.org/https://doi.org/10.1016/j.ejor.2003.10.039
  • Hand David, J., & Heikki, M. (2001). Principles of data mining (adaptive computation and machine learning). Bradford Books.
  • Hasin, M. A. A., Ghosh, S., & Shareef, M. A. (2011). An ANN approach to demand forecasting in retail trade in Bangladesh. International Journal of Trade, Economics and Finance, 2(2), 154.
  • KARADAVUT, U., Aşır, G.E.N.Ç., TOZLUCA, A., KINACI, İ., AKSOYAK, Ş., PALTA, Ç. and PEKGÖR, A (2005). Nohut Cicer arietinum L. Bitkisinde Verime Etki Eden Bazı Karakterlerin Alternatif Regresyon Yöntemleriyle Karşılaştırılması. Journal of Agricultural Sciences, 11(03), 328–333.
  • Lakshmi Anusha, S., Alok, S., & Shaik, A. (2014). Demand Forecasting for the Indian Pharmaceutical Retail: A Case Study. Journal of Supply Chain Management Systems, 3(2).
  • Liu, L.-M., Bhattacharyya, S., Sclove, S. L., Chen, R., & Lattyak, W. J. (2001). Data mining on time series: an illustration using fast-food restaurant franchise data. Computational Statistics \& Data Analysis, 37(4), 455–476.
  • Makridakis, S., Wheelwright, S. C., & Hyndman, R. J. (2008). Forecasting methods and applications. John wiley \& sons.
  • Martin, L.-C., & others. (2019). Machine learning vs. traditional forecasting methods An application to South African GDP.
  • Merkuryeva, G., Valberga, A., & Smirnov, A. (2019). Demand forecasting in pharmaceutical supply chains: A case study. Procedia Computer Science, 149, 3–10.
  • Micajkova, V., Georgieva Svrtinov, V., Petkovski, V., & Esmerova, E. (2018). Selecting appropriate forecast method on the basic of forecast accuracy-pharmaceutical company case study.
  • Moskowitz, H. (1972). The Value of Information in Aggregate Production Planning—A Behavioral Experiment. A I I E Transactions, 4(4), 290–297. https://doi.org/10.1080/05695557208974865
  • Padhy, N., Mishra, D., Panigrahi, R., & others. (2012). The survey of data mining applications and feature scope. ArXiv Preprint ArXiv:1211.5723.
  • Pavlyshenko, B. M. (2019). Machine-learning models for sales time series forecasting. Data, 4(1), 15. Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo, M., & Rigol-Sanchez, J. P. (2012).
  • An assessment of the effectiveness of a random forest classifier for land-cover classification. ISPRS Journal of Photogrammetry and Remote Sensing, 67, 93–104.
  • Smits, R. E. H. M., & Boon, W. P. C. (2008). The role of users in innovation in the pharmaceutical industry. Drug Discovery Today, 13(7–8), 353–359.
  • Tosun, T. (2006). Veri madenciliği teknikleriyle kredi kartlarında müşteri kaybetme analizi Fen Bilimleri Enstitüsü.
  • Tugay, R., & Oguducu, S. G. (2020). Demand Prediction Using Machine Learning Methods and Stacked Generalization. arXiv. https://doi.org/10.48550/ARXIV.2009.09756
  • Yıldırım, N. (2010). En küçük kareler, Ridge regresyon ve Robust regresyon yöntemlerinde analiz sonuçlarına aykırı değerlerin etkilerinin belirlenmesi.. Çukurova Üniversitesi Fen Bilimleri Enstitüsü, Yüksek Lisans Tez, Adana.
There are 28 citations in total.

Details

Primary Language English
Subjects Engineering
Journal Section Research Articles
Authors

Salih İmece This is me 0000-0002-7162-6559

Ömer Faruk Beyca 0000-0002-0944-6813

Early Pub Date September 30, 2022
Publication Date September 30, 2022
Published in Issue Year 2022 Volume: 34 Issue: 3

Cite

APA İmece, S., & Beyca, Ö. F. (2022). Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry. International Journal of Advances in Engineering and Pure Sciences, 34(3), 415-425. https://doi.org/10.7240/jeps.1127844
AMA İmece S, Beyca ÖF. Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry. JEPS. September 2022;34(3):415-425. doi:10.7240/jeps.1127844
Chicago İmece, Salih, and Ömer Faruk Beyca. “Demand Forecasting With Integration of Time Series and Regression Models in Pharmaceutical Industry”. International Journal of Advances in Engineering and Pure Sciences 34, no. 3 (September 2022): 415-25. https://doi.org/10.7240/jeps.1127844.
EndNote İmece S, Beyca ÖF (September 1, 2022) Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry. International Journal of Advances in Engineering and Pure Sciences 34 3 415–425.
IEEE S. İmece and Ö. F. Beyca, “Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry”, JEPS, vol. 34, no. 3, pp. 415–425, 2022, doi: 10.7240/jeps.1127844.
ISNAD İmece, Salih - Beyca, Ömer Faruk. “Demand Forecasting With Integration of Time Series and Regression Models in Pharmaceutical Industry”. International Journal of Advances in Engineering and Pure Sciences 34/3 (September 2022), 415-425. https://doi.org/10.7240/jeps.1127844.
JAMA İmece S, Beyca ÖF. Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry. JEPS. 2022;34:415–425.
MLA İmece, Salih and Ömer Faruk Beyca. “Demand Forecasting With Integration of Time Series and Regression Models in Pharmaceutical Industry”. International Journal of Advances in Engineering and Pure Sciences, vol. 34, no. 3, 2022, pp. 415-2, doi:10.7240/jeps.1127844.
Vancouver İmece S, Beyca ÖF. Demand Forecasting with Integration of Time Series and Regression Models in Pharmaceutical Industry. JEPS. 2022;34(3):415-2.