当前位置: X-MOL 学术Comput. Econ. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Menu Optimization for Multi-Profile Customer Systems on Large Scale Data
Computational Economics ( IF 1.9 ) Pub Date : 2021-07-24 , DOI: 10.1007/s10614-021-10147-0
Jeyhun Karimov 1 , Murat Ozbayoglu 2 , Bulent Tavli 2 , Erdogan Dogdu 3
Affiliation  

Everyday, a majority of the people, most probably several times, use the banking applications through online applications or physical ATM (Automated Teller Machine) devices for managing their financial transactions. However, most financial institutions provide static user interfaces regardless of the needs for different customers. Saving even a few seconds for each transaction through more personalized interface design might not only result in higher efficiency, but also result in customer satisfaction and increased market share among the competitors. In ATM Graphical User Interface (GUI) design, transaction completion time is, arguably, one of the most important metrics to quantify customer satisfaction. Optimizing GUI menu structures has been pursued and many heuristic techniques for this purpose are present. However, menu optimization by employing an exact mathematical optimization framework has never been performed in the literature. We cast the ATM menu optimization problem as a Mixed Integer Programming (MIP) framework. All the parameters of the MIP framework are derived from a comprehensive actual ATM menu usage database. We also proposed two heuristic approaches to reduce the computational complexity. Our solution can be accustomed with ergonomic factors and can easily be tailored for optimization of various menu design problems. Performance evaluations of our solutions by using actual ATM data reveal the superior performance of our optimization solution.



中文翻译:

大规模数据上多资料客户系统的菜单优化

每天,大多数人(很可能多次)通过在线应用程序或物理 ATM(自动柜员机)设备使用银行应用程序来管理他们的金融交易。然而,无论不同客户的需求如何,大多数金融机构都提供静态用户界面。通过更加个性化的界面设计,每笔交易甚至可以节省几秒钟,不仅可以提高效率,还可以提高客户满意度并增加竞争对手的市场份额。在 ATM 图形用户界面 (GUI) 设计中,交易完成时间可以说是量化客户满意度的最重要指标之一。一直在追求优化 GUI 菜单结构,并且存在许多用于此目的的启发式技术。然而,在文献中从未使用精确的数学优化框架进行菜单优化。我们将 ATM 菜单优化问题视为混合整数规划 (MIP) 框架。MIP 框架的所有参数都来自一个综合的实际 ATM 菜单使用数据库。我们还提出了两种启发式方法来降低计算复杂度。我们的解决方案可以适应人体工程学因素,并且可以轻松定制以优化各种菜单设计问题。通过使用实际 ATM 数据对我们的解决方案进行的性能评估揭示了我们优化解决方案的卓越性能。MIP 框架的所有参数都来自一个综合的实际 ATM 菜单使用数据库。我们还提出了两种启发式方法来降低计算复杂度。我们的解决方案可以适应人体工程学因素,并且可以轻松定制以优化各种菜单设计问题。通过使用实际 ATM 数据对我们的解决方案进行的性能评估揭示了我们优化解决方案的卓越性能。MIP 框架的所有参数都来自一个综合的实际 ATM 菜单使用数据库。我们还提出了两种启发式方法来降低计算复杂度。我们的解决方案可以适应人体工程学因素,并且可以轻松定制以优化各种菜单设计问题。通过使用实际 ATM 数据对我们的解决方案进行的性能评估揭示了我们优化解决方案的卓越性能。

更新日期:2021-07-24
down
wechat
bug