Containerizing Smart Requirements

Engineering a Lifeboat for the New Normal

By | 4 minute read | September 22, 2020

Requirements are the DNA for projects ranging from straightforward software development to the complexities of engineering, including product development with Internet of Things (IoT) and “systems of systems.” We know this. Automation, analytics and processes that improve requirements consistency, relevance, adaptability and resilience increase business agility and can cuts costs dramatically. And the earlier potential problems are identified and addressed, the more positive the impact.

Issues discovered later in the development process cost 200x or more to remediate.

Yet, effective requirements strategies are more drastically needed now more than ever. We are moving through a myriad of crises, including a global pandemic and associated economic, business and resource volatility, climate change and related life-critical weather disasters, spawning new compliance demands, and geopolitical and social turmoil worldwide. The chaos of these combined challenges overwhelm standard approaches. They demand leverage of advanced technologies and analytics such as Machine Learning (ML) and Artificial Intelligence (AI) along with requirements automation and agile processes. A range of flexible application deployment options including hybrid cloud and containerized alternatives create easier access for distanced collaboration while addressing management, administration and privacy concerns.

Current disruption to engineering and development teams includes not merely a dramatic (and often immediate) shift to digitization and remote work (while dealing with supply chain interference), but also staff reductions in the wake of economic unpredictability. At the same time, business responsiveness to these circumstances of crisis falls on the backs of software and engineering teams who galvanize product and competitive execution.

Working with clients during this disruption has given me a deep appreciation for the optimism and fortitude of these teams who are the “tip of the spear” for many companies forging strategies for survival and success in our current world, where we obviously can take nothing for granted. So what are technical automation and other resources to empower these leaner, dispersed groups?

As we move into the “new normal,” scaffolding a requirements, development and engineering strategy in combination with ML and AI and established industry standards, such as the International Council on Systems Engineering (INCOSE) Guide for Writing Requirements, can benefit organizations in significant ways:

  1. Leveraging AI in conjunction with INCOSE early on can help users identify and assess requirements that are set up poorly to enable teams to address those issues and to increase efficiency upfront;
  2. Team reviews benefit from identifying ambiguous, incomplete and poorly structured requirements. Remediating these factors as soon as possible saves time, cuts costs and help create a solid product and project foundation to be able to pivot rapidly to address changing market needs as we recover;
  3. Less experienced engineers have an opportunity with these combined capabilities to vet their initial requirements for straightforward violations or inconsistencies so that reviews by senior staff can focus on sophisticated requirements needs and challenges;
  4. Availability of combined automated requirements and smart analytics solutions on a range of platforms – from cloud to on-premises to containers – enables flexible adoption to encompass needs for distributed access as well as privacy and risk concerns.

Organizations can find it challenging to effectively deploy engineering software modules across servers, linking and tuning these servers and administering that environment. Container implementations can address this complexity and lower barriers of entry for companies seeking to embrace and deploy comprehensive development management solutions. Increased efficiency is especially important as organizations deal with resource constraints, including less staff.

In that context, IBM announced its Engineering Requirements Quality Assistant (RQA) on Red Hat’s Open Shift and is shipping the product as of September 22, 2020. RQA uses IBM’s Watson AI in conjunction with DOORS Next Generation to assess and rank requirements quality using INCOSE’s Guide for Writing Requirements and provides in-line help on how to address existing problems with consistency, structure, etc.

This is significant as IBM deploys its first engineering product on an OpenShift container and also delivers the first iteration of Watson AI-powered engineering for private cloud consumption. Before RQA on OpenShift, customers had to send their requirements to the cloud if they wanted to leverage Watson AI to evaluate their requirements, which was problematic for some customers with IP, privacy and other concerns. RQA on OpenShift allows customers to deploy the Watson AI-powered RQA in their own data centers.

I am pleased to see combined product availability for IBM and Red Hat, which has taken some time. As a first step, this sets the stage for additional IBM ELM solutions becoming available in OpenShift containers.

IDC recommends the following next steps:

  1. Leverage this challenging time to assess your company’s development tools, requirements and processes to prioritize areas on which to focus and evolve an effective, adaptive strategy;
  2. To address high complexity demands, evaluate ML and AI capabilities to use in conjunction with requirements and other areas such as quality and security (as the technology becomes available) to benefit from intelligent analytics along with industry guides such as INCOSE’s;
  3. Use analytics to make metrics actionable and pragmatic, determining and addressing root-cause issues rather than mere symptoms;
  4. Evaluate and adopt solutions that free up resources from mundane tool maintenance and integration for a comprehensive engineering lifecycle approach that benefits from smart analytics, agile processes and flexible deployment options such as hybrid cloud and containers;
  5. Be able to pivot – position yourself, your teams and your requirements to be rapidly responsive to our unpredictable “new normal” with end to end visibility of value chains, assessing of data input relevance and of model assumptions and output;
  6. Champion effective AI and ML adoption with tools, training, expertise, standard setting and guidelines to enable and encompass “trust” and adaptability.

Read: Adopting Engineering Lifecycle Management Tools to Combat Market Uncertainty