Scholarlink

<p>Ensuring the gasoline supply while minimizing procurement costs presents a significant challenge in the face of uncertainties in procurement prices and consumer demand. These challenges are amplified by the price commitment condition, which greatly affects consumer behavior. Moreover, the absence of historical procurement price and demand data, further complicates the formulation of effective procurement strategies. To address these challenges, this study proposes a Two-stage Self-adaptive Online Procurement Method (TSOPM) to optimize gasoline procurement under uncertainties from prices and demand. In the warm-up stage, procurement decisions are guided by daily procurement prices, working capital, inventory volume, and accumulated historical data, which serve to train demand and price prediction models. Specifically, the demand prediction model is tailored to consider the impact of price fluctuation discrepancies on demand under price commitment conditions. In the steady stage, outputs from the prediction models are integrated into robust optimization methods to ascertain the procurement quantities under uncertainties from prices and demand. To further enhance solution quality, a Dual-population online procurement Dandelion Optimizer (DDO) algorithm is developed. Empirical results confirm the effectiveness of the proposed model, demonstrating superior performance in demand prediction accuracy and procurement cost reduction when compared to existing advanced methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Two-Stage Procurement Strategy Under Price Commitment: Integrating Self-adaptive Online Learning with Robust Optimization

  • 期刊论文

  • 发表日期：

    2026年3月16日

查看全文
  • Boyang Li,
  • Yunzhe Qiu,
  • Yilan Shen,
  • Xi Zhang

摘要

Ensuring the gasoline supply while minimizing procurement costs presents a significant challenge in the face of uncertainties in procurement prices and consumer demand. These challenges are amplified by the price commitment condition, which greatly affects consumer behavior. Moreover, the absence of historical procurement price and demand data, further complicates the formulation of effective procurement strategies. To address these challenges, this study proposes a Two-stage Self-adaptive Online Procurement Method (TSOPM) to optimize gasoline procurement under uncertainties from prices and demand. In the warm-up stage, procurement decisions are guided by daily procurement prices, working capital, inventory volume, and accumulated historical data, which serve to train demand and price prediction models. Specifically, the demand prediction model is tailored to consider the impact of price fluctuation discrepancies on demand under price commitment conditions. In the steady stage, outputs from the prediction models are integrated into robust optimization methods to ascertain the procurement quantities under uncertainties from prices and demand. To further enhance solution quality, a Dual-population online procurement Dandelion Optimizer (DDO) algorithm is developed. Empirical results confirm the effectiveness of the proposed model, demonstrating superior performance in demand prediction accuracy and procurement cost reduction when compared to existing advanced methods.