Author(s)

Junxiang Wang ¹, Lingyu Dai ², Wenxiao Xie ¹

Affiliation(s)

¹ School of Economics, Beijing Technology and Business University, Beijing, China
² College of Economics and Management, Northeast Agricultural University, Harbin, Heilongjiang Province, China

Corresponding Author

Junxiang Wang

ABSTRACT

The rapid expansion of China’s pharmaceutical e-commerce sector, accelerated by post-pandemic digitalization and supportive policies, presents critical operational challenges for market players. This study leverages 2020–2021 transactional data from TMall’s vitamin supplement market to develop actionable strategies for optimizing online pharmacy operations. Through Python and Excel-driven preprocessing—standardizing discount rates (null→1) and calculating sales (price × quantity × discount)—we analysed 1.8 million records spanning 26 stores, 9,697 products, and 465 brands. Our analysis revealed significant market concentration: Ali Health Pharmacy dominated 45% of total sales, with immune-boosting products driving peak revenue during seasonal campaigns. The top 10 brands (led by elevit and FANCL) collectively captured 65% of vitamin sales, primarily targeting women, children, and seniors through wellness-positioned supplements. For forecasting, a dual-model approach was implemented. While Grey Prediction GM(1,1) projected Q1 2022 vitamin sales at ¥105.23 million, Winters' additive time series model (α=0.109, β/γ≈0) demonstrated superior accuracy (R²=0.95, MAPE=5.3%), refining the estimate to ¥87.75 million by capturing subtle cyclical patterns overlooked by conventional methods. These findings inform three evidence-based strategies: First, forging partnerships with integrated health platforms (e.g., Ali Health) enhances consumer trust and regulatory navigation. Second, prioritizing high-margin supplements like prenatal vitamins and calcium boosters aligns with demographic-driven demand. Third, AI-powered dynamic pricing—such as bundled discounts during low-season months (June/October)—can smooth revenue volatility. The 18.7% forecast error reduction achieved through time-series modeling underscores its value for inventory planning in volatile pharmaceutical markets, particularly as telehealth integration reshapes global healthcare delivery.

KEYWORDS

Pharmaceutical E-Commerce, Sales Forecasting, Inventory Optimization, Time Series Analysis, Health Supplement Market

CITE THIS PAPER

Junxiang Wang, Lingyu Dai, Wenxiao Xie, Predictive Analytics for Pharmaceutical E-Commerce: Leveraging Grey Forecasting and Time Series Models to Optimize Vitamin Sales Strategies on TMall. Social Medicine and Health Management (2025) Vol. 6: 103-114. DOI: http://dx.doi.org/10.23977/socmhm.2025.060114.

REFERENCES

[1] Jovic A, Frid N, Brkic K, et al.Interpretability and accuracy of machine learning algorithms for biomedical time series analysis – a scoping review[J]. Biomedical Signal Processing and Control, 2025,110(PB): 108153.
[2] Rezaei Z, Samghabadi S S, Amini A M, et al.Predicting climate change: A comparative analysis of time series models for CO2 concentrations and temperature anomalies[J]. Environmental Modelling and Software, 2025, 106533.
[3] Song L, Li Z, Zhang J, et al. An hourly and localized optimization method for soil fugitive dust emission inventory based on machine learning[J]. Journal of Environmental Sciences, 2024.
[4] Abuella M, Fanaee H, Nowaczyk S, et al. Time-series analysis approach for improving energy efficiency of a fixed-route vessel in short-sea shipping[J]. arXiv preprint arXiv:2402.00698, 2024.
[5] Iezzi F D, Monte R. E-word of mouth in sales volume forecasting: Toyota Camry case study[J]. Big Data Research,2025, 100542.
[6] Jing J, Lu C, Luwen Z, et al.A study on the factors that influence consumers’ continuance intention to use artificial intelligence chatbots in a pharmaceutical e-commerce context[J]. The Electronic Library, 2025,43(3):303-321.
[7] Mondal R, Manna K A, Akhtar M, et al. Optimization of a price break policy and advertisement effort based non-instantaneous deteriorating inventory problem with partial backlogged via metaheuristic algorithms[J]. Mathematics and Computers in Simulation, 2025, 236: 221-247. 
[8] Mahmudah A N, Park K Y, Ock M. Trends of health behaviors among cancer patients in Korea: a cross-sectional time series analysis[J]. BMC Cancer, 2025, 25(1):1124.
[9] Jiang W, Zhao M, Li H. Time series image coding classification theory based on Lagrange multiplier method[J]. Scientific Reports, 2025, 15(1):20697. 
[10] Angel D M, Nunes M, Peacock O, et al. Analysing longitudinal wearable physical activity data using non-stationary time series models[J]. International Journal of Behavioral Nutrition and Physical Activity, 2025, 22(1):88.
[11] Jia J, Chen L, Wu C, et al. User avoidance behavior in pharmaceutical e-commerce intelligent customer service: a stressor-strain-outcome perspective [J]. Frontiers in Psychology, 2025, 16: 1514571.
[12] Wang P, Chen M, An W, et al. Research on the fugitive soil dust emission inventory in Western China based on wind erosion equation parameter optimization[J]. Frontiers in Environmental Science, 2023, 11: 1301934.

Sponsors, Associates, and Links

---

• Information Systems and Economics
  
  
• Accounting, Auditing and Finance
  
  
• Industrial Engineering and Innovation Management
  
  
• Tourism Management and Technology Economy
  
  
• Journal of Computational and Financial Econometrics
  
  
• Financial Engineering and Risk Management
  
  
• Accounting and Corporate Management
  
  
• Social Security and Administration Management
  
  
• Population, Resources & Environmental Economics
  
  
• Statistics & Quantitative Economics
  
  
• Agricultural & Forestry Economics and Management
  
  
• Land Resource Management
  
  
• Information, Library and Archival Science
  
  
• Journal of Human Resource Development
  
  
• Manufacturing and Service Operations Management
  
  
• Operational Research and Cybernetics