Designing for Reliability: Crafting Resilient Products and Systems for Tomorrow

In 1971, one of the oldest and most well-known American luggage brands came up with one of the most hard-hitting product demonstrations of all time. The ad-campaign featured an angry gorilla trying to vandalize a piece of luggage and putting its resilience to the ultimate test. Needless to say, the campaign helped the brand successfully highlight the product’s only reliability quotient—durability.

It wasn’t long ago that the main selling point of products was their own intrinsic features like durability, appearance and capability. Today however, products are no longer standalone, instead they are connected and can communicate with each other. Even a modern counterpart of the carry-on luggage bag comes embedded with high-tech features like Bluetooth-controlled smart lock, a USB port for charging, and GPS-enabled tracking.

The inherent complexity of modern day products, in terms of hardware and software architecture is fast transforming the way a company defines a product’s reliability requirements. The focus is more on laying down stringent quality and testing parameters for each component, since a minute functional error within the circuitry or embedded software applications is enough to put the reliability of a product at stake. The immediate results of this may be anything ranging from consumer dissatisfaction to penalties worth billions of dollars as warranty infringement compensation.

The need of the hour thus is to commit resources for a robust design for reliability (DFR) process that can be implemented right from the concept stage of product development to the obsolescence stage. That way, businesses can minimize failures, ensure maintainability, and shorten repair times in an environment where no one gets a second chance to make an impression.

Infusing Reliability into the Heart of a Product

Stringent testing and debugging cycles during the early phase of the product development process are often considered to be fail-proof methods of ensuring that a shipped product is reliable. But a rigorously tested product can apparently be stable, yet still lack reliability while functioning. For instance, a software application for an anti-lock brake system in a car might appear reliable, but will definitely not function if one of the input sensors starts malfunctioning.

The key to holistically address such reliability issues across the entire product lifecycle might very well lie in developing systems that are steadfast even in the most adverse run-time situations. Such systems should be able to address component failures as they occur, so that the magnitude of repercussions are reduced.

One way through which product developers can achieve such high levels of reliability is by eliminating all single points of failure with redundancy engineering – a fault tolerant architecture already implemented successfully in the avionics industry. With this method, a redundant unit can be added around a critical system to serve as a backup. For the anti-lock brake system of a car, this could mean installing a second sensor in line with the first one.

‘Fail-safe’ mechanisms are another critical consideration for product reliability. An effective yet low-cost alternative to redundant systems, failure assessment mechanisms like design failure mode and effect analysis (DFMEA) can help designers determine the root cause of each potential failure point. With this reliability testing method, numerical metrics can be assigned for each factor that product developers can then leverage to implement fail-safes, redundancies, and other components. Ultimately, the system’s overall reliability index can be increased.

Fault Tree Analysis (FTA) and Highly Accelerated Life Testing (HALT) are the two other effective failure analysis methods that can help reliability engineers identify discrepancies within a system. Fault tree analysis is a methodical deductive failure analysis process where a particular system fault is broken down into several individual causes. Each of these immediate causes are then resolved until the root cause of a failure is determined. On the other hand, HALT is a more directive failure analysis approach that exposes a product to diverse stress points to discern the physical drawbacks of the design.

Such meticulous reliability engineering methods can have a substantial impact across industries like avionics, medical, automotive and industrial process where safety-critical products and systems are a mandate. Making a product fault-proof at the design and manufacturing stage can even lower the chances of product recall, which often leads to catastrophic outcomes for companies. A new report reveals that a serious recall incident can cost a company upwards of $12 million, a price too high to pay for an average company.

Designing Fail-Proof Products of the Future

Technologies like augmented reality (AR) and virtual reality (VR) add a whole new dimension to the concept of product and system reliability analysis. Product developers can leverage these technologies to simulate real-world scenarios and test the reliability of products ranging from crash barriers and aircraft wings to pipeline components and car safety applications.

There are organizations that have already started walking in that direction by developing targeted reliability analysis solutions that are based on virtual simulation. A renowned automated test equipment and virtual instrumentation software provider recently launched an advanced vehicle radar test system (VRTS), which can help automotive engineers simulate real-world scenarios like pedestrians walking along roads or cars changing lanes. This application can be used to test and configure advanced driver assistance systems (ADAS) that can eventually enhance the reliability and safety of autonomous vehicles.

As a natural consequence, virtual reliability testing simulations can help deliver significant cost savings during the iterative design phase itself. With this in the arsenal, product designers can curb design time and reduce the costs of developing physical models for tests and successive modifications. Additionally, these testing simulations can help software developers achieve product reliability standards by offering 100% traceability and reporting coverage for all possible end-user scenarios (positive, negative, failures).

There’s no denying that going forward, the race to develop technologically sound products and systems will intensify for one reason alone — customer satisfaction will belong to only those who plan their game with reliability as a trump card.