基于智能优化算法和多决策模型的界面布局优化算法

Interface layout optimization method based on intelligent optimization algorithm and MCDM

本研究从提高人机界面可用性、优化界面布局的角度出发,提出了一种综合智能优化算法、熵权法-层次分析法(AHP)-逼近理想解排序法(TOPSIS)的界面布局优化方法。首先,以计算机断层扫描仪界面的高保真原型,为实验载体进行眼动追踪实验,获取到界面元素之间的关联度;建立界面布局优化的目标函数和多约束条件,采用蚁群算法(ACO)对目标函数进行求解,获得备选界面布局方案;从界面的效率、有效性、满意度3个方面将鼠标点击次数、鼠标轨迹长度、错误数量、任务持续时间和SUS问卷分数作为评价指标。采用熵权法、AHP和TOPSIS形成的多准则决策(MCDM)模型评估并选出最优界面布局方案。实验结果表明,最优界面布局方案TOPSIS综合得分为4.948037,与原始界面相比,完成任务所需的鼠标点击次数减少了11.90%、鼠标轨迹长度减少了35.60%、错误数量减少30.43%、任务持续时间减少了26.89%、SUS分数增加了2.75%。本研究使用的方法获得的最佳界面布局方案,有效地提高了人机界面的可用性。

In this study,from the perspective of improving the usability of human-computer interface and optimising the interface layout,an interface layout optimisation method integrating intelligent optimisation algorithm,entropy weight method-hierarchical analysis method(AHP)-approximation of ideal solution sorting method(TOPSIS)is proposed.Firstly,a high-fidelity prototype of the interface of the computed tomography scanner is used as an experimental carrier for eye-tracking experiments to obtain the correlation degree between the interface elements;the objective function and multiple constraints of the interface layout optimization are established,and the Ant Colony Algorithm(ACO)is used to solve the objective function and obtain the alternative interface layout scheme;from the three aspects of the interface,namely,the number of mouse clicks,the length of mouse track length,number of errors,task duration and SUS questionnaire score as evaluation indexes from three aspects of interface efficiency,effectiveness and satisfaction.The optimal interface layout was evaluated and selected by using the multi-criteria decision making(MCDM)model formed by entropy weight method,AHP and TOPSIS.The experimental results show that the optimal interface layout scheme has a TOPSIS composite score of 4.948037,which reduces the number of mouse clicks required to complete the task by 11.90%,reduces the length of the mouse track by 35.60%,reduces the number of errors by 30.43%,reduces the duration of the task by 26.89%,and increases the SUS scores by 2.75%compared to the original interface.The optimal interface layout solution obtained from the methodology used in this study effectively improved the usability of the human-computer interface.