Model-driven development platforms shift the focus of software development activity from coding to modeling for enterprises. A significant number of such platforms are available in the market. Selecting the best fitting platform is challenging, as domain experts are not typically model-driven deployment platform experts and have limited time for acquiring the needed knowledge. We model the problem as a multi-criteria decision-making problem and capture knowledge systematically about the features and qualities of 30 alternative platforms. Through four industry case studies, we confirm that the model supports decision-makers with the selection problem by reducing the time and cost of the decision-making process and by providing a richer list of options than the enterprises considered initially. We show that having decision knowledge readily available supports decision-makers in making more rational, efficient, and effective decisions. The study’s theoretical contribution is the observation that the decision framework provides a reliable approach for creating decision models in software production.

模型驱动的开发平台将软件开发活动的重点从企业的编码转移到了建模。市场上有大量此类平台可用。选择最合适的平台具有挑战性，因为领域专家通常不是模型驱动的部署平台专家，并且获取所需知识的时间有限。我们将问题建模为多准则决策问题，并系统地获取有关30个替代平台的功能和质量的知识。通过四个行业案例研究，我们确认该模型可以通过减少决策过程的时间和成本以及提供比最初考虑的企业更丰富的选择列表来为决策者提供选择问题的支持。我们表明，拥有随时可用的决策知识可以支持决策者做出更合理，更有效的决策。该研究的理论贡献是观察到决策框架为在软件生产中创建决策模型提供了可靠的方法。