Steinbeis student analyzes the automation level of a software release process
Improving the efficiency, speed, cost-effectiveness, and, above all, the quality of business and IT processes is increasingly becoming a key factor for future competitiveness. Automating tasks, individual steps within processes, or even complete processes – including across departmental boundaries – is now of central importance and not only beneficial for companies, but also for their employees. This was also something ascertained by Smart Solutions for Industry employee Arne Steckler, as part of his business competence project for his Master of Science degree in Business Information Systems at Steinbeis University. For his project, he helped Steinbeis Interagierende Systeme (a supplier of software used on a platform that tests driver assistance systems) to document its software release process based on suitable methods – also analyzing the potential to automate and thus optimize processes. The process he looked at is used by the company to perform merges, ensure there are no errors, and deliver new software versions.
Automated processes save time and are often more reliable and less prone to errors. As a result, not only do they facilitate significant productivity enhancements, they can also lead to major cost savings. Both of these aspects are crucial for quality assurance processes. Whereas humans are clearly superior to machines when it comes to flexible or creative tasks, machines are almost invariably better and much faster with mindless, repetitive tasks. With automated processes, it is entirely possible to ensure quality assurance tasks are completed and carried out correctly, but not all tasks can be performed by computers, and sometimes they require or even wholly necessitate the creativity and flexibility of a human being. Automation levels are thus an important indicator of a company’s process maturity. Automating processes is a good way to achieve all three corners of the magic time, quality, and cost triangle, especially when people are given the best possible support: Ultimately, being faster (time) and achieving fewer errors (quality) also makes things less expensive (costs).
A win-win situation for companies and workers
Automation does not just offer advantages to companies, however; employees also derive benefit in a number of areas, because it is rarely about cutting resources. More often, it is about avoiding mind-numbing and repetitive tasks in order to free people up to engage in more valuable work that requires a variety of human skills. In essence, available time can be put to better use and made more interesting for employees. Furthermore, enabling highly automated processes to dovetail with tasks performed by humans offers ways to ensure that defined (internal or statutory) processes and reporting standards are adhered to, as well as compliance regulations.
A step-by-step process leading to success
The objective of the project conducted by Arne Steckler on behalf of Steinbeis Interagierende Systeme was to determine the degree of automation required within a software release process and pinpoint potential to make improvements when it comes to automation. On the one hand, this involved documenting the current standard of the process, thus making it transparent for management and people responsible for the process. On the other, Steckler was to highlight tasks offering the strongest potential to be automated. It was important to pay particularly close attention to specific areas in which human steps should be kept in place, also showing how they could be systematically dovetailed and safeguarded.
Steckler was also asked to provide a clear overview of the process in the form of a BPMN model highlighting communication interfaces. This would offer a better, shared understanding of the process and result in an overview of the process in the form of documentation.
To kick off the project, Steckler conducted one-on-one interviews with people involved in the process to discuss any of their tasks that overlapped with the chosen process. Their responses were recorded, documented, and transferred into a BPMN model. To ensure information was captured correctly and to rule out misunderstandings, the documented process was reviewed and refined in follow-up meetings with the interviewees. Once all process tasks and relevant artifacts (model elements) had been successfully captured, a sign-off meeting was organized to allow all participants in the process to validate the process documentation as an overall model.
The very procedure of capturing processes and providing a clear representation of them in BPMN is an opportunity to make improvements – simply by encouraging stakeholders involved in the process to actively think about how it works, to talk to one another, and to see the process in its entirety in order to point to potential in visible terms. Such models can also provide input further down the line for additional, in-depth assessments. Such assessments can be used to capture the process in value terms, also highlighting weak points, the potential to make improvements, and the causes of potential discrepancies between actual and target process performance.
For Steckler, the focus of his process analysis lay in the degree of automation of the software release process. His aim was to ascertain the extent to which certain tasks had already been automated and to pinpoint areas in which there was meaningful potential to convert manual tasks into automated routines. Overall, there should be further reductions in manual tasks and, in the future, it should be possible to produce automated reports.
The first task for Steckler was to develop a formula for calculating automation levels, taking the overall circumstances into account as described above. Based on this, he defined options for evaluating the potential to automate individual steps. The factors required to calculate automation levels were provided by stakeholders responsible for carrying out the process. In addition, these factors were used to identify tasks offering the best potential to automate processes.
Leveraging documentation and transparency to improve quality
Documenting the release process for the first time created a common understanding among all process stakeholders, also making it possible to work together in pinpointing initial areas to make optimizations and provide a user-friendly representation of the process. In addition, the project participants have added transparency to the automation level of the process, identified further potential for the company to automate processes, and made it possible to assess automation levels in numerical terms. Implementing the identified improvements will improve the quality of artifacts and thus enhance produced software. In follow-up interviews, staff gave positive feedback on the approach, particularly regarding the role they played as the starting point for process improvements, and the fact that the quantitative assessment showed in visible terms how they contributed in their role to automation.