"Zero Defects" has long been used as a management term aimed at eliminating defects in industrial production e.g. - as a consumer, if you buy a high-end smartphone, tablet, or a smart television, you expect them to run flawlessly without any defect throughout the expected lifetime of the gadget. Zero defects was conceived as a way of thinking or a philosophy to "do it right the first time" aided with processes and tools to achieve the objective.
Today, enterprises are exploring the possibilities to extend the same concept to IT application implementations and operational support throughout the lifecycle of the application. The concept is interestingly framed as an ambition to move towards "Zero Incidents". Is this a realistic ambition or to reframe this conundrum giving it a positive twist - How can this be achieved and to what extent?
Let us now examine the two critical stages in the lifecycle of an application contributing to incidents:
- The cycle starts with the implementation of an application via an IT project. This is the stage which sets up the direction, an application will take through its lifecycle. Inadequate analysis of the business requirements and evaluation of technical design may lead to future problems and incidents.
- The second stage is when the business users start using the application once implemented. They expect the application to run as per the process design which has been finalized to fulfil the business requirement. The problems or the incidents occur when the application become unavailable to the business user probably due to capacity related issues or they are not functioning in line with the business requirement it was designed for.
Extensive analysis of the business requirements, evaluation of design alternatives, scalability of the application considering growth in future usage, adoption of appropriate coding standards and rigorous testing are some of the key levers for arriving at a near perfect design leading to zero incidents. While this can be a detailed discussion topic for another blog, we will focus our discussion on the approach that can be adopted to accomplish our ambition of reaching “Near Zero Incidents” in the post implementation - application support phase.
Disruptions in business-as-usual to the application user occurs either when (a) the application is unavailable for usage due to capacity issues, (b) the application is not functioning, as it was originally intended to or it is not functioning in the most optimal or efficient manner, (c) when the business user is not very conversant with the system & need some help on how-to kind of issues. The ambition to drive towards "Zero Incidents" regime would need an approach which addresses all the above three dimensions.
Embracing the philosophy of "Near Zero Incidents" requires a holistic approach addressing the key dimensions discussed above and developing / deploying processes and tools (discussed below) to accomplish the stated ambition.
- Setting up an Operational Command Center (OCC) - Running SAP operations like a factory is an extension of the OCC concept and means operating efficiently, systematically, and by exception - in a controlled yet agile business environment. The OCC approach towards management of operations include collecting both Application related & Business Process related monitoring information, converting any inconsistencies as alerts & proactively managing the alerts before they cause disruption to the environment.
- Problem Management deploying a HotSpot Analysis Technique - HotSpot Analysis is a technique of monitoring "critical business processes" end-to-end & identifying problem issues within these processes. This technique allows identification of the root cause of a problem by looking at the entire process chain and analyzing dependent tasks which might lead to the problem incidents. This enables finding a more comprehensive solution for a problem which may include fixing weak links in a process thereby maximizing process availability and business value from the application.
- Business Process Visualization - Complex processes and multiple variations of the same process causes the business to slow down and run at a sub-optimal level. These also become hotbeds for business dissatisfaction and future disruptions. Business Process Visualisation helps in identifying these processes that can be simplified & harmonised aided with Process Mining tools and sets up the foundation for transforming these processes.
- Artificial Intelligence triggering Self-Help / Self-Healing Capabilities - Self-Help / Self-Healing solutions developed around the principles of Robotic Process Automation (RPA) for repetitive tasks requiring minimal decision making interventions and Cognitive Automation where the solutions learn from past occurrences and takes corrective actions. These solutions when deployed and exposed to business users can aid in Self-Help as well as Self-Healing capabilities.
The blog here attempts to provide an overview of the various processes that can be adopted to drive towards a "Near Zero Incident" regime. Let me know what you think on the approach outlined.