Schedule a Software Walkthrough
Requs AI Training Series

Predict software failure rate, availability, and reliability

These predictions tell you early whether the reliability and availability requirements can actually be met — with enough time left to pursue an alternative path if they can't.

predict software failure rate availability downtime training RAM metrics
1 Day
Virtual, self-guided — work through it at your own pace
2 Prerequisites Required
Reliable Software 101, and Predict Software Defects and When They Will Be Discovered
Recommended For
Teams using Requs AI Predict — the software itself isn't required to take the class
01 — Authority

Our founder is the global leader in software reliability prediction

Standards Body

IEEE 1633 Leadership

Chair of the 2026 IEEE 1633 working group — the governing standard for software reliability engineering.

Continual lessons learned applied to the world's largest defect density benchmarking study

Decades of Trending Data

While others fail to keep their model factors current with technology - we're on our 8th major revision since 1993.

World's largest database of software failures analyzed by root cause

CDE Taxonomy inventor

From it, we created the Common Defect Enumeration, the primary taxonomy currently adopted and cited across DOW technical frameworks.

02 — The Class

Find out early whether the requirement can be met at all

Reliability and availability requirements are usually treated as something to verify at the end of a program — after the architecture is locked, after the budget is spent. This class teaches how to predict failure rate, availability, and reliability early enough that if the requirement can't be met, there's still time to change course.

This class directly supports the DoW Reliable Software SOW task Reliable Software Predictions, and pairs naturally with Requs AI Predict — but doesn't require owning it. It's built to be useful on its own.

Prerequisites Required This class requires completing Reliable Software 101 and Predict Software Defects and When They Will Be Discovered first — this class builds directly on both.
Capability
Without this training
With this training
Predicting software failure rate
No structured method — guesswork or hardware analogies
A repeatable method, taught step by step
Predicting software availability
Not modeled until the system is already deployed
Modeled before a line of code is written
Predicting software downtime
Discovered the hard way, once it's already happened in operation
Estimated and budgeted for in advance
Standards alignment
Not tied to any recognized standard
Maps directly to IEEE 1633 clause 5.3
how to predict software failure rate availability downtime RAM metrics

Reliable software predictions, broken down: predicted MTBSF and P(success) roll up into RAM metrics; predicted downtime and availability roll up into SLA/RAM metrics; and predicted defects, arrival rate, pileup, time to first occurrence, and risk level roll up into SQA and test metrics. This class covers how each of these predictions is actually made.

03 — Curriculum

What the class covers

Three modules, covered in a single self-guided day.

§1Failure Rate

How to predict software failure rate

What You'll Learn

How to predict software failure rate using the same reliability growth foundations covered in the prerequisite classes, applied specifically to a failure-rate requirement.

Why It Matters

Failure rate is the number most reliability requirements are actually written against — get comfortable predicting it, not just reporting it after the fact.

how to predict software failure rate trend chart

Failure rate predictions in practice: predicted MTTCF, unavailability, and unreliability trending month over month, with upper and lower bounds shown as the model updates with new data.

§2Availability

How to predict software availability

What You'll Learn

How to translate a predicted failure rate and repair time into a predicted availability number — before the system exists to measure it directly.

Why It Matters

Availability requirements are usually the ones programs get surprised by late. Predicting it early removes the surprise.

§3Downtime

How to predict software downtime

What You'll Learn

How to predict expected downtime — both frequency and duration — from the same underlying failure rate and repair data used in the earlier modules.

Why It Matters

Downtime is what a customer actually experiences. A failure rate number alone doesn't tell you what that looks like operationally — predicted downtime does.

What's Next Once you've completed this class, the series continues with System Availability Objectives & Error Budgets, Integration with Hardware Predictions, and How to Navigate IEEE 1633 Clause 5.3 — later classes in the Requs AI Training Series.
04 — Class Details

Format, duration, and prerequisites

Duration

1 Day

Delivered as a single day of self-guided instruction.

Format

Virtual, Self-Guided

Work through the material on your own schedule, at your own pace.

Prerequisites

Reliable Software 101 & Predict Software Defects

Both required before taking this class. Requs AI Predict itself is recommended but not required to own.

Find out whether the reliability requirement is achievable — before it's too late to change course.

Register for the class, or schedule a demonstration of Requs AI Predict.