February 1, 2017
By Pravin Asar, Millenium Engineering and Integration
Successful new product introduction (NPI) is key to business success. The consumer is king and drives the success or failure of the corporations. In today’s globally connected economy, offering a tailored product portfolio that can cater to the ever-changing preferences of both local and global markets means organizations need to have a clear understanding of customer requirements in different segments, and must design products that meet those expectations. Incorporating the ever-changing requirements into a typical phased-gate/stage-gate product development is a challenge.
Adding to the challenge is the cutthroat competition in the global market. Today, the race is not just about creating new products, but doing so faster than the competition. The ability to decrease time to market, improve first time right performance and improve cost competitiveness, are emerging as key differentiators.
A successful NPI process must use current and past records such as sales/marketing, design, manufacturing operations and testing/service data to develop new products. Doing so offers huge potential to improve product performance, bring about efficiency in the product development process, contain costs and enhance customer experience.
For example, product ideas, customer behavior patterns, voice of customer (VoC) data, quality function deployment (QFD) data, and product trends from social networks and listening platforms, can help design the product strategy and portfolio. Warranty, quality and testing data, data from CAD systems and manufacturing process data can help in product design and validation, as information is fed back into the NPI process.
This data is not available in a simple format, but is generally in large volumes, and is dispersed across the enterprise. In order to benefit from this big volume of constantly evolving data, organizations must have a well-defined strategy to collect, store, synthesize and disseminate it in the form of knowledge required for various business functions.
Key Definitions: Data, Information and Knowledge
The terms data, information and knowledge are sometimes used interchangeably, but are in fact three distinct concepts as shown in Fig. 1.
Data consists of unprocessed discrete and objective facts about events, properties of objects, etc. Data is mostly structured and has no value unless processed and analyzed.
Information is derived by the aggregation and analysis of data. It is used mostly to assist in decision making by answering questions related to who, what, when and how many. Information is a message, usually in the form of a document or audio-visual communication. As with any message, it has a sender and a receiver. Information is meant to change the way the receiver perceives something, and to have an impact on their judgment and behavior.
Knowledge is know-how and is what makes possible the transformation of information into instructions. Knowledge can be obtained either by transmission from another who has it; by instructions (explicit knowledge); or by extracting it from experience (tacit knowledge). Essentially, knowledge is evaluated and organized information that can be used purposefully in the problem solving process.
The core of the product development process lies in knowledge and its reuse. Knowledge—when managed effectively—can help reduce NPI project time, improve product quality and increase customer satisfaction. In a knowledge-based organization, it plays a crucial role in guiding the organization’s actions and establishing a sustainable competitive advantage. Typically, the NPI process uses a combination of data, information and knowledge.
Wisdom, the ability to increase effectiveness by judicial use of the knowledge base, comes from the expertise and insight of a few individuals in the organization, and is usually manifested as policies, best practices and lessons-learned artifacts.
Making Knowledge Management Integral to Product Development
Four kinds of “know” have been distinguished by Organization for Economic Co-operation and Development (OECD), which are know-what, know-why, know-how and know-who.
Know-what refers to knowledge about facts. Know-how refers to scientific knowledge of the principles and laws of nature. Know-why refers to skills or the capability to do something. Know-who involves information about who knows what and who knows how to do what.
We could represent engineering knowledge creation during new product development (NPD) by posing the following questions:
• What product knowledge is created or represented?
• Who are the actors playing what roles in creating, using or modifying product knowledge?
• Where is the product knowledge created and located?
• How is the product knowledge being created or modified?
• Why was certain product knowledge created or modified?
• When was the product knowledge created or modified?
The conventional product development process (Fig. 2) starts from strategizing the product to designing, validating and testing it, and finally ends with the product being phased out.
The concept of integrating and sharing heterogeneous knowledge emerged in the late 1990s, when product lifecycle management (PLM) systems were conceptualized and knowledge–based engineering began attracting attention. Earlier PLM systems were essentially product data management (PDM) systems. More recent PLM systems now attempt to manage many more aspects of product engineering. However, they still do not provide end-to-end tools for product development.
Emerging Trends in Product Development
NPD teams are increasingly becoming cross-functional organizational groups, dependent on internal as well as external stakeholders. Additionally, crowdsourcing, social media, collaboration, use-of-field data, capture technology and additional tools are being utilized for product idea screening and refinement. Fiat Mio and Procter & Gamble (P&G) Products are notable examples of successful collaborative product development.
Obviously, significantly more data resides outside the realm of engineering data. During the product development lifecycle, large volumes of several types of data are generated throughout the design, analysis, testing and service operations of the products. All of this data is available to be processed and used as a reusable knowledge base, and to provide real time intelligence to the product development teams. In most cases, data such as customer insights (structured and unstructured), competitive intelligence and product performance data can prove to be of great value for the overall success of both the new generation of the product and future generations.
Knowledge Management and Big Data
Knowledge management is getting the right information to the right people at the right time. Furthermore, it is helping team members create, share and act upon that knowledge in ways that will measurably improve the performance of an organization and its partners (Fig. 3).
Big Data analysis is normally associated with censuses, surveys and similar applications. It can be used to positively affect the product lifecycle, as a strategic initiative to meet the goals of decreasing time to market, improving first-time-right performance, and improving cost competitiveness. The four dimensions of Big Data a.k.a. the 4Vs—volume, velocity, variety and value—are also applicable to new product development. These four Vs are elaborated upon below.
• Volume: Machine-generated data is produced in much larger quantities than non-traditional data.
• Velocity: This refers to the speed of data processing, or the latency rate at which analytics must be applied to the data, and looped back to the original sources of data to action. Social media data streams are not as massive as machine-generated data and produce a large influx of opinions and relationships valuable to customer relationship management.
• Variety: This refers to the large variety of input data (customer insights, competitive intelligence, trends, benchmarking data, standards and materials) that in turn generate a variety of data (CAD/CAM/CAE data, drawings, documents, test data, product and process performance data) as output.
• Value: The economic value of different data varies significantly depending upon both the source and its end use.
A Big Data strategy requires the ability to sense, acquire, transmit, process, store and analyze data that generates knowledge and can be stored in a repository for later use. To derive maximum benefits from Big Data Analysis and Data Mining, organizations must make sure their IT infrastructure supports the 4Vs, with the goals of knowledge capture (rapid rate of delivery), knowledge representation (extraction of huge volumes of data, with varying data types), knowledge transfer (availability of knowledge), and knowledge reuse.
Big Data in Action
The use of Big Data and its analysis in product development is mindfully driven by the available technologies in the organization and the tight integration between hardware and software and other data generation mechanisms. An approach to analyze the use of Big Data and the consequent knowledge in NPI is shown in Fig. 4.
A typical NPD process interfaces with enterprise-wide systems such as ERP, CRM and PLM, to retrieve information. The automobile industry is currently leading the pack in this regard with their integration of the Internet of Things (IoT) to capture field data. This information is explicit and structured. Information and data are exchanged on a continuous basis with these systems as the product is being realized.
Although a small part of this information flows back into the enterprise systems, attempts should be made to acquire this in a central repository, typically a single data warehouse, to capture data of the new product initiative for future needs. A deliberate attempt must be made to keep the data together so that it can be combined to create information that can then be analyzed to generate the knowledge repository and to support future product development processes.
The key benefit of a centralized authoritative and controlled data-driven approach is the ability to use a single source of data that drives many decisions concurrently and reduces data duplication. The mantra “create once, use everywhere” should be followed. For example, continuous availability of market data, product planning data and VOC data can be used to improve the features and designs of current and future products.
A service-oriented architecture (SOA) coupled with real-time data analysis is beneficial in production capacity, facility planning and to stay ahead of potential capacity expansion problems. For example, a warranty or a field issues database will alert product designers to product issues that can be eliminated while a batch is still in production. Real-time automobile performance data from sensors pertaining to engine performance or driver behavior can help product developers capture potential performance issues or add new features to the vehicle.
Product development is a stage-gate and cross-functional process, dependent on many internal and external data and practices. Big Data usage in product development offers new ways to integrate data with product design and process improvement. Having a SOA in place for real-time data analytics can derive optimal benefits. Some notable recommendations are listed below.
• Set a clear vision for knowledge management in your organization.
• Perform a thorough assessment of your current status, pain points and future requirements.
• Devise a well thought out longterm strategy for knowledge management.
• Prepare a roadmap to achieve the vision and successful strategy implementation.
• Implement knowledge management as a business transformation program that aligns with your organization’s business processes, technology, infrastructure and change management practices.
Pravin Asar is a principal systems engineer at Millenium Engineering and Integration. He has over 20 years of experience in CAD/CAM and PLM software development and integration with enterprise systems. Send email about this article to [email protected]