When dealing with consensus cost problems with asymmetric adjustment costs, the uncertain scenarios with certain probabilities which are becoming a serious problem decision-makers have to face. However, existing optimization-based consensus models have failed to consider uncertain factors that could influence the final consensus and total consensus cost. In order to better deal with these issues, it is necessary to develop practical consensus optimal models. Thus, we establish three two-stage stochastic minimum cost consensus models with asymmetric adjustment costs that may eventually lead the way to better consensus outcomes. The impact of uncertain parameters (such as individual opinions, unit asymmetric adjustment costs, compromise limits, cost-free thresholds) are investigated by modeling three kinds of uncertain consensus models. We solve the proposed two-stage stochastic consensus problem iteratively using the L-shaped algorithm and show the convergence of the algorithm. Furthermore, a case of pollution control negotiations verifies the practicability of the proposed models. Moreover, the comparison of results with the L-shaped algorithm and CPLEX shows that the L-shaped algorithm is more effective in solving time. Some discussions and comparisons on local and global sensitivity analysis of the uncertain parameters are presented to reveal the features of the proposed models. Finally, the relationships between the minimum cost consensus model and minimum cost consensus models with asymmetric adjustment costs and the proposed models are also provided.

在处理具有不对称调整成本的共识成本问题时，具有某些概率的不确定场景将成为决策者面临的严重问题。但是，现有的基于优化的共识模型未能考虑可能影响最终共识和总共识成本的不确定因素。为了更好地处理这些问题，有必要开发实用的共识最优模型。因此，我们建立了三个具有不对称调整成本的两阶段随机最小成本共识模型，这些模型最终可能会导致更好的共识结果。通过对三种不确定性共识模型进行建模，研究了不确定性参数（如个人意见，单位非对称调整成本，折衷限制，无成本阈值）的影响。我们使用L形算法迭代地解决了所提出的两阶段随机共识问题，并证明了该算法的收敛性。此外，污染控制谈判的案例验证了所提出模型的实用性。此外，与L形算法和CPLEX的结果比较表明，L形算法在求解时间上更有效。对不确定参数的局部和全局敏感性分析进行了一些讨论和比较，以揭示所提出模型的特征。最后，给出了最小成本共识模型与具有不对称调整成本的最小成本共识模型之间的关系以及所提出的模型。