Design for Reliability (DfR) identifies key design parameters and potential failure rates inherent in a design and develops procedures to ensure that a product meets its reliability requirements for the duration of its lifetime. Digital Transformation is impacting product reliability. Trends in product personalization and connected devices make DfR increasingly important in today’s market. Unfortunately, according to LNS Research, most manufacturers lack the reliability technology strategy and top management sponsorship to have successful reliability programs.
Successful reliability engineering requires the ability to predict what portions of a product may fail as well as the performance, safety, and economic impacts of failure. It requires transparency to current and past issues to improve predictions, react to failures, and continuously improve. Successful DfR must be supported by effective product management strategies.
With the rise of the Internet of Things (IoT) and centralized data management platforms like Product Lifecycle Management (PLM) technology, companies can create a closed-loop, data-driven DfR program to improve predictability and reliability and achieve better-performing products.
DfR’s Top Challenges
According to a survey performed by LNS Research, the top four challenges to DfR success are lack of top management sponsorship, late engagement in New Product Introduction (NPI), siloed processes and technology, and poor predictability. Furthermore, research reveals that manufacturers primarily manage DfR data with spreadsheets and electronic documents.
- 90 percent of the companies surveyed manage requirements in spreadsheets and electronic documents
- 83 percent manage FMEAs (Failure Modes and Effects Analysis) in spreadsheets and electronic documents
- 70 percent manage new product test data in spreadsheets and electronic documents
Like many product design and development processes, this disconnected, document-centric approach is a significant obstacle to the analytics needed to generate accurate historical reliability statistics and lessons learned. It also creates silos among DfR teams and processes, which reduces cross-functional engagement and creates cross-discipline process silos, and ultimately results in late DfR engagement.
This is where PLM technology and the IoT come into play. Together, they can help address top DfR challenges by automating processes, improving visibility/access to key data, and engaging DfR early in the NPI process.
Engaging DfR Early on with PLM
PLM technology provides a central location to manage all product information within a controlled environment and provides the platform to plug in reliability and IoT data. Reliability teams tend to engage late in the product development cycle, after the Bill of Material (BOM) is finalized. By leveraging PLM as their data platform, reliability engineers can gain better visibility into product information among all teams and engage earlier in the design process/NPI stage. PLM can be used to tie information from various sources to improve DfR such as IoT data and incidents and issues from test labs. This information can be fed back to engineering and other key departments, including the c-level, via PLM.
For example, in this real-world scenario, a design engineer at a medical device company kept designing in the same sensor part with a high failure rate. Because they did not have a central system to track and manage field-level data, Engineering did not have visibility into this issue and the effect the sensor had on the product. Did the failure trigger a service call? What was the sensor component costing the company? By giving engineers access to data on how products perform in the market with IoT-integrated PLM, their product specifications can include data beyond design engineering such as reliability, quality and marketing.
IoT is a Double-Edged Sword
IoT data often has the ability to deliver data points in real time, removing much of the traditional ambiguity around product performance and usage. Product planning, design, and quality departments can learn from a product’s operational behavior via IoT to improve features that customers use most. Product quality issues, failures in the field, software bugs, and/or customer feedback data that is fed into PLM enables manufacturers to track and configure product design requirements based on usage patterns and allow for the redesign of parts or systems to improve quality. IoT devices leverage the technology to report data to collection and analysis systems. This data can also be integrated into PLM systems to provide data directly to the product record. Issues and tolerance breaches from IoT devices can also initiate support and quality processes in the PLM system.
The IoT can be a double-edged sword for reliability. It creates product complexity, which inherently poses a reliability challenge, but also delivers much greater insight into equipment usage. This insight enables predictive maintenance for existing products, which improves availability. By capturing actual usage and other conditions leading to failure, it also yields data that can greatly improve next-generation products. Used properly, the IoT can result in substantial gains in product reliability and availability.
As an example, a global manufacturer of medium, heavy-duty, and severe-service trucks started to monitor fleets remotely as a service. Through that work, the firm found "canaries," or early failures that ultimately indicate a broad issue with the fleet, similar to canaries in a coal mine. The truck manufacturer identifies a pattern in sensor data by re-viewing the events and data leading up to a component failure, then automatically checks other vehicles to see if the same pattern exists.
This era, which promises real-time product feedback, is a sea change for engineering and there are many great possibilities. Historically, engineers create product designs from a set of requirements then hand off the finished model to manufacturing, which manages production processes with its own systems and tools. Field service and maintenance personnel typically have another set of data related to the product, housed in a separate system, and few of the disciplines are likely to share information, let alone work off a common data set.
An open PLM platform can integrate with IoT applications to capture data, which triggers automated workflows and alerts to address problems. PLM becomes a platform to push and pull information for helping identify issues early in the design process.
Closed-Loop, Data-Driven DfR
As mentioned previously, today’s spreadsheet and electronic document approach is the origin of many downstream DfR challenges such as late engagement in NPI and poor predictability. According to a recent report from LNS Research, What’s Your Design for Reliability (DfR) Data Plan?, LNS recommends a closed-loop, data-driven approach to DfR that establishes a formal, change-managed process connected to PLM, the NPI system of record. This method provides connectivity to gather risk and reliability experiences from across multiple business and operational systems into a centralized model. It is important to organize these experiences and related metrics against data objects that are relevant and re-usable to DfR, including parts, configurations, and serial numbers. This closed-loop, data driven approach ensures better accuracy and visibility into prior product issues, and comparison of predicted to experienced reliability for the current product. This approach also enables product development teams to engage DfR more fully, including it in their development decisions. It allows DfR teams to participate earlier in the lifecycle, and also creates a closed-loop connection between virtual/physical test, DfR, and product development.
DfR and the Bottom Line
DfR can contribute to the bottom line in many ways. Improvements in warranty costs and product return/recall rates, customer satisfaction, and safety can all result from a successful DfR program. It is clear that the market has changed and along with it the importance of reliability. Trends in IoT/connected devices and product personalization will require manufacturers to put new processes and technology in place to accurately capture, analyze and address issues. Companies should look at leveraging their PLM platform, the system used to develop their product, as a source to connect other systems/data into like DfR and IoT to create a holistic view of the product and support data-driven processes.
Chuck Cimalore is CTO at Omnify Software.
