铁路站房改造项目主材供应商履约评价与动态淘汰机制研究 

摘要：铁路站房改造工程往往在既有线路持续运营的条件下推进，具有工期高度压缩、施工界面交叉密集、外部干扰强等特征，钢结构、幕墙铝板、装饰石材、机电管线、防水材料等主材的供应履约水平直接决定改造进度与工程质量。然而，当前工程主材供应商管理仍多依赖一次性招标准入与年度定性考核，缺乏对履约全过程的动态量化评价与优胜劣汰机制，低效供应商难以被及时识别和退出，进而形成质量与工期隐患。针对这一问题，本文从质量、进度、成本、服务协同、安全环保五个维度出发，构建了包含 18 项二级指标的主材供应商履约评价体系；采用层次分析法（Analytic Hierarchy Process，AHP）刻画专家经验、熵权法提取数据客观差异，并按组合系数进行主客观组合赋权，再结合逼近理想解排序法（Technique for Order Preference by Similarity to Ideal Solution，TOPSIS）计算各供应商的综合履约贴近度；在此基础上，设计以 “滑动考核窗口 + 动态淘汰阈值 + 预警 — 淘汰 — 递补三级规则” 为核心的动态淘汰机制，实现供应队伍的持续优化。基于铁路站房改造场景构造的仿真算例（15 家供应商、6 个考核期、90 条履约记录）表明：组合赋权将质量与进度维度合计权重由 AHP 单一赋权的 0.612 回调，并使成本、服务权重回升，有效抑制了单一赋权偏倚；TOPSIS 贴近度介于 0.004 至 0.959，呈现清晰梯队，末位三家供应商与主体拉开显著差距；动态机制在连续 6 期内稳定淘汰末位低效供应商，在册队伍平均履约得分由 0.606 提升至 0.768，其中幸存供应商自身整改带来约 10.0% 的相对提升，叠加末位淘汰与递补的结构性优化，队伍整体均分累计提升约 26.7%；灵敏度分析进一步揭示，考核窗口不小于 3 期且淘汰阈值系数适中（0.85–0.90）时，可在及时淘汰真实末位与避免误伤中游供应商之间取得平衡。研究可为铁路站房改造工程主材供应商的全周期量化管理提供决策支持。

关键词：铁路站房改造；主材供应商；履约评价；组合赋权；TOPSIS；动态淘汰机制

正文内容 / Content：

可下载并阅读全文PDF，请按照本文版权许可使用。

Download the full text PDF for viewing and using it according to the license of this paper.

Performance Evaluation and Dynamic Elimination Mechanism for Main-Material Suppliers in Railway Station Renovation Projects

Abstract: Railway station renovation projects are usually carried out while existing lines remain in operation, and are therefore characterized by tight schedules, dense construction interfaces and strong external interference. The delivery performance of main materials—such as steel structures, curtain-wall aluminum panels, decorative stone, electromechanical pipelines and waterproofing materials—directly determines the renovation progress and engineering quality. However, current supplier management still largely relies on one-off tendering admission and annual qualitative appraisal, lacking dynamic quantitative evaluation and a survival-of-the-fittest mechanism over the whole performance process, so that inefficient suppliers cannot be identified and removed in time. To address this problem, an evaluation index system with 18 secondary indicators covering five dimensions—quality, delivery, cost, service-and-coordination, and safety-and-environment—is constructed. The Analytic Hierarchy Process (AHP) captures expert experience, the entropy weight method extracts objective data differences, and the two are fused by combination weighting; the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is then used to compute each supplier's comprehensive closeness. On this basis, a dynamic elimination mechanism centered on a sliding assessment window, a dynamic elimination threshold and a three-level "warning–elimination–replacement" rule is designed. A simulated case with 15 suppliers, 6 assessment periods and 90 performance records shows that combination weighting suppresses single-weighting bias, that TOPSIS closeness ranges from 0.004 to 0.959 and forms clear tiers, and that the mechanism steadily eliminates bottom-ranked suppliers while raising the average performance score of retained suppliers from 0.606 to 0.768—about 10.0% from survivors' own improvement plus a structural gain from elimination and replacement, for a cumulative increase of about 26.7%. Sensitivity analysis further reveals that an assessment window of at least 3 periods with a moderate threshold coefficient (0.85–0.90) balances timely elimination against mis-elimination.

Keywords: railway station renovation; main-material supplier; performance evaluation; combination weighting; TOPSIS; dynamic elimination mechanism

参考文献/References

1. Kutlu Gündoğdu F, Kahraman C. A novel spherical fuzzy analytic hierarchy process and its renewable energy application[J]. Soft Computing, 2020, 24(6): 4607-4621.
2. Wang Y, Xu Z. Evaluation of the human settlement in Lhasa with intuitionistic fuzzy analytic hierarchy process[J]. International Journal of Fuzzy Systems, 2018, 20(1): 29-44.
3. Chen P. Effects of the entropy weight on TOPSIS[J]. Expert Systems with Applications, 2021, 168: 114186.
4. Chakraborty S. TOPSIS and Modified TOPSIS: a comparative analysis[J]. Decision Analytics Journal, 2022, 2: 100021.
5. Wang Y, Liu P, Yao Y. BMW-TOPSIS: a generalized TOPSIS model based on three-way decision[J]. Information Sciences, 2022, 607: 799-818.
6. 施明泰，罗以谨，李欣. 国家电网供应商绩效评价体系研究[J]. 电气时代，2018.
7. 罗冰洁，刘建，金蓓，等. 医用耗材供应商绩效评价指标体系研究[J]. 中国数字医学，2018.
8. 余敢华，李菡婷. 基于模糊综合评判法的超市生鲜食品供应商绩效评价[J]. 萍乡学院学报，2019.
9. 方沅湘，贾昱晞，周伟，等. 基于供应商绩效评价体系的体外诊断试剂精细化管理[J]. 中国卫生产业，2021.
10. 叶华，谢卫华，卜祥磊，等. 基于平衡计分卡的医用耗材供应商绩效管理探讨[J]. 中国医疗设备，2019.
11. 刘立国，梁炳南. 基于熵权-TOPSIS 模型的液压实验台供应商选择评价[J]. 机床与液压，2020.
12. 王一雷，朱庆华，夏西强. 基于模糊 AHP-GP 的低碳供应商选择模型[J]. 运筹与管理，2020.
13. 郭子雪，张运通，田雨，等. 基于直觉模糊多属性决策的逆向物流供应商选择[J]. 河北大学学报（自然科学版），2021.
14. 姚罗瑞，马从安. 基于直觉模糊 TOPSIS 法的煤炭企业设备供应商选择研究[J]. 数学的实践与认识，2020.
15. 王江华，朱雪欣，宋兆龙. 基于界面管理的装配式建筑供应链集成风险识别与评价[J]. 青岛理工大学学报，2020.
16. 张逢雨，谢俊燕. 基于云模型的装配式建筑供应链风险评价[J]. 智能建筑与智慧城市，2020.
17. 孙亚琪，田源. 基于复杂网络理论的装配式建筑供应链关键风险研究[J]. 建筑经济，2020.
18. 闫文周，王莹，冯中帅. 基于区块链的建筑供应链信息共享演化博弈研究[J]. 科技管理研究，2021.
19. 汤萍. 关于铁路企业物资供应商动态管理的探讨[J]. 经营管理者，2011.
20. 王明磊. 供应商绩效管理综合评价模型研究[J]. 商场现代化，2020.