The goal of successful Product Lifecycle Management (PLM) is to support all processes and services related to products for their entire lifetime. The lifecycle phases of a physical product are beginning of life (BOL), comprising design and manufacturing of a product, middle of life (MOL), containing its use and maintenance, and finally end of life (EOL), including reuse, recycling, and disposal of the product or its components. Currently, the big gap of information between enterprise applications and the world of physical products makes it difficult for PLM to anticipate and deal with changes of products in a timely and effective manner.
For the majority of today’s technological products and especially for those producing “hi-tech” waste – consumer electronics, household “white” machines, vehicles – it is fair to say that the information flow breaks down after the delivery of the product to the customer, when leaving the BOL phase. First, there is a lack of proper process models that address a closed loop information flow for the entire product life cycle. Innovative business models must be developed to provide product owners with an incentive to reveal information about the usage or condition of their assets. Second, the hardware and software technology allowing to access, maintain, and utilize product data in a ubiquitous way must partially be developed and standardized to find wide market acceptance.
Closing the information loop
As a member of the PROMISE initiative (PROduct lifecycle Management and Information tracking using Smart Embedded systems), SAP Research helps to develop a new generation of PLM solution that uses smart embedded IT devices to allow the seamless flow of product data through its complete lifecycle. All actors in a product’s lifecycle should be able to manage and control this product information at any moment of its lifecycle at any place in the world. A look at the real-world scenarios of some market leaders in manufacturing involved in the PROMISE project help understand the need for closing the current information loop between different lifecycle phases of products.
Bombardier for example designs and produces over 400 different types of locomotives. The aim of the global leader in rail equipment and servicing is to close the information loop between the experience in service, like field data, and the knowledge needed in order to develop improved locomotives for specific criteria, such as, design for reliability, availability and maintainability or life cycle costs, design for product safety and design for environment. Based on field data collected during the operation of locomotives, both proven designs and design issues can be identified for reuse and investigation, respectively.
Caterpillar at the other hand, the world market leader in manufacturing of construction and mining equipment, purposes to support the decommissioning of heavy-load machinery at the end of its life. Instead of manual inspection of a component’s status, a monitoring system is envisioned to systematically collect data during the machines operation. When the machine is decommissioned, the data associated with the built-in parts is retrieved and is combined with data on the economic demand for decision making. The appropriate handling of the various components can be determined, such as disposal, recycling, reuse, or remanufacturing of components in order to increase part usage.
Last but not least, Fiat seeks new ways to better understand the customer habits and the mission profile of Iveco commercial vehicles, to improve the effectiveness of fleet management. The objective is to provide customers with flexible maintenance planning, which is based on the actual degradation of vehicle components instead of fixed intervals. The correct timing for maintenance is determined by measuring the wear-out of selected critical components with sensors that are integrated into the vehicle. To estimate actual machine degradation and to allow for customized maintenance plans, sensor data, for example oil consumption, vibrations or engine speed, are to be captured and analysed. With this approach, the customers can increase their return on assets – as costly breakdowns are avoided while preventing the replacement of parts which are still in good condition.
The semantics and structure of that data are more complex than that of identification data, and must be specified in advance to allow for correct interpretation and further processing, like calculating averages and trends. The embedded processing units of complex products like cars, trucks or locomotives typically utilize proprietary software and communication protocols, which have to be addressed at a proper abstraction layer. What’s more, enterprise assets, such as locomotives, vehicles and machines, are typically mobile, and may be used and serviced in a wide geographical area.
To identify unique products
For all these needs PROMISE exploits product-embedded information devices (PEIDs), in order to access product-specific information at any point of time within the products’ lifecycle. PROMISE enables both device management and communication between one or more PEIDs and existing enterprise backend software. The networking layer of the PROMISE-PLM system as a whole is build as middleware, enabling different players in the network to have controlled and secure access to relevant information. This middleware layer is a key element that allows connection of the phases of life and closing the loop. According to requirements, information can be stored on-board on the PEID, locally at the site of the owner of the information or where the information has been created, or it can be stored in a centralised data warehouse. PEIDs are mainly based on new ubiquitous computing techniques, for instance RFID, sensor networks, onboard computers, with the capability of wireless communication. These devices ensure the availability of up-to-date product-specific information during the complete lifetime of the product.
Providing enterprise applications with up-to-date data the bottom layer of the high-level architecture of PROMISE consists of physical products involved in PLM, such as cars, locomotives, and trucks. Identification and operational data, for example current mileage, fuel consumption, and average engine temperature, from such products can be collected by PEIDs either on a regularly basis or upon request. The PROMISE middleware communicates with the PEIDs to obtain the data and it further filters the data in order to deliver useful information to the actors involved in PLM. The middleware should be able to identify those objects and communicate with their PEIDs to exchange status data.
Finally, in enterprise applications, the information is processed using business intelligence algorithms and transformed into meaningful knowledge on products, such as the expected remaining lifetime for a product based on its measured conditions. Laws and regulations concerning decommissioning of products lead to further challenges for the capture and management of product data. For example, product or part information must be kept available for time periods of up to 20-30 years for construction machinery and locomotives. During such periods, the products will cross company borders, or may reside with one or more customers. This raises both security and privacy issues to be dealt with, which are also being investigated by SAP Research together with partners in the range of the project PROMISE.
Seamless integration of information devices
Although taking part in designing and developing all PROMISE architecture components like business processes, applications, PEIDs and products, the major focus of SAP Research lays on the middleware. It aims is to enable PLM applications to access to complete and up-to-date product information regardless of the current location and usage of the product and of the PEID technology employed. The major challenges in developing such a middleware are to overcome the heterogeneity of product data and PEID technologies and to achieve a high robustness, scalability, and transparency of the deployment in a distributed environment.
The strategy of SAP to achieve these objectives is to develop a modular architecture consisting of dedicated components for device handling and request handling, respectively, which can be easily distributed over different locations. The device handling component utilizes an Universal Plug and Play (UPnP) -based protocol for automatic recognition of PEIDs and communication with them. The request handling component provides a uniform web-service interface for backend applications to query data from PEIDs. Besides basic read and write operations, an extensible library of data preprocessing services for converting, aggregating and filtering the data, such as the creation of average, evaluation of extremal values or sampling, is supported so that data reduction can be early performed within the middleware. These services are available for all applications and the library are expandable with new services. Most of the functionality of the middleware like device registration, authentication, communication management or inquiry processing, is generic, making it usable for different enterprise applications, for example mySAP SCM and mySAP PLM processes such as tracking and tracing of products, fleet management, or industrial facilities maintenance.