Review
Product intelligence in industrial control: Theory and practice

https://doi.org/10.1016/j.arcontrol.2013.03.003Get rights and content

Abstract

This paper explores the evolving industrial control paradigm of product intelligence. The approach seeks to give a customer greater control over the processing of an order – by integrating technologies which allow for greater tracking of the order and methodologies which allow the customer [via the order] to dynamically influence the way the order is produced, stored or transported. The paper examines developments from four distinct perspectives: conceptual developments, theoretical issues, practical deployment and business opportunities. In each area, existing work is reviewed and open challenges for research are identified. The paper concludes by identifying four key obstacles to be overcome in order to successfully deploy product intelligence in an industrial application.

Introduction

In an industrial context, the notion of product intelligence describes the linking of a physical order or product to information and rules governing the way it is intended to be made, stored or transported. Thus, an approach that follows the product intelligence paradigm for industrial control can be described as one that has the ability to treat each individual product instance differently depending on its specific characteristics and needs. The industrial process being controlled might be production, logistics or maintenance. In a product intelligence approach, each part/product/order is capable of collecting and storing information about its special requirements which can be unique for each instance. This product-instance level information can then be used both for static (planning) and dynamic (execution) processes although it is often more logically linked to dynamic ones. Moreover, a product intelligence approach might involve products which seek to satisfy their own needs and achieve their special goals through transactions with their service providers rather than leaving the providers to choose how they should be handled. Hence, it provides a particular form of supervisory control where the product or order can be responsible for triggering different processing steps.

This is illustrated with a simple logistics example in Fig. 1 in which a violin to be shipped from a supplier to a customer is directly associated not only with transportation details but with a set of guidelines for adjusting delivery date and route in the event of a delay. The information and rules might for example represent the needs or intentions of the customer with regard to how the product might be managed by supplier organisations. If it could be effectively automated, such a facility then enables the product – and hence indirectly the customer – to be directly represented within a supply organisation’s information systems environment. Furthermore, if the automation extended to managing the decisions associated with the violin’s shipment, then the order begins to be able to influence its own logistics. In a similar manner, Fig. 2 illustrates a production order generating an intelligent software agent to “accompany” a car body through production. Here, an intelligent product negotiates an assembly path with different production resources (Chirn & McFarlane, 2000) – achieving so called product driven control.

The notion of product intelligence was introduced in 2002 when several of the authors presented an alternate vision for the way in which supply chains might work (McFarlane et al., 2003, Wong et al., 2002). At the same time other researchers were working on a similar approach, see, for example, (Kärkkäinen, Holmström, Främling, & Artto, 2003). This work described supply chain operations in which parts, products or orders (i.e. collections of products) could monitor and potentially influence their own progress through the industrial supply chain. At this time, issues of development and wide-scale adoption of the internet and also RFID (Radio Frequency Identification) technologies were receiving significant attention (Sarma, 2001). The supply chain model based around product intelligence provided a conceptual focus for these developments. Also, from an operational perspective, the model promised the potential for greater flexibility and versatility, although at the time more emphasis was placed on improving efficiencies and supply chain visibility and reducing costs in an increasingly global supply chain environment.

Following from the previous section, one clear motivation for a product intelligence based approach is to provide a customer with the opportunity to inform and influence the supply of his orders. Although not explicitly mentioned yet, there is an assumption that this “influencing” is done in an automated manner. This is particularly appropriate for situations where the specific priorities of the customer (for a particular order) differ from the overall priorities of a manufacturer or logistics provider. In Table 1 a number of such situations have been identified to illustrate this. In each case the needs or interests of the order (or the order owner) might differ from those of the supplier handling the order. These situations are grouped into static scenarios – those relating to the structure of the situation and dynamic scenarios – those relating to operational variation.

These scenarios highlight situations where the nature of the processing of the product might change, where the customer and supplier have different perspectives or priorities or where the supplier may have a limited ability to access information or influence the progress of an order. Hence we might refer to product intelligence as providing an order or customer-oriented approach to managing operations as opposed to a more conventional organisation-oriented approach. We will discuss this interpretation further in the next section.

In addition to motivating Product Intelligence as a paradigm for providing an automated customer-oriented approach to industrial control, the aims of this paper are two-fold. Firstly, to provide a structured, critical review of developments in the area and secondly to identify and present the open questions and challenges that need to be studied before product intelligence can be deployed in real industrial systems.

The following sections are divided (rather artificially) into conceptual issues (Section 2), theoretical developments and challenges (Section 3), practical issues in deployment (Section 4) and business challenges and barriers (Section 5). Because one of the roles of the paper is to overview developments, there is no single literature review section, but rather existing literature is referenced in each section as relevant. The paper concludes with a forward agenda for the field.

Section snippets

Product intelligence: concepts

In this section we gather together definitions, characteristics and terminologies that have been associated with product intelligence since the early 2000s. We also tie this work into other academic developments which have common features. More specifically, the section attempts to answer the following questions relating to the product intelligence concept:

  • How can an intelligent product be defined?

  • What is a product intelligence approach for industrial control?

  • How do the different available

Product intelligence: theory

Developing a control system with the characteristics described in Section 2.4 brings with it specific challenges in both the design and subsequent analysis of such systems. These arise both from the desire to produce an automated and distributed solution and also the need to provide for external influence on what would otherwise be external control operations. This latter challenge is illustrated in Fig. 4 in which the active nature of the provider’s operations is contrasted with the passive

Product intelligence: practice

We now turn our attention to examining issues and challenges in deploying product intelligence in practice. We begin with a discussion on industrial systems that already use some features of product intelligence and we then review similar deployments found in the literature. We then summarise the main open issues around the deployment of systems following an intelligent product approach in real cases. Finally, following the structure of our previous sections, we close this one with a set of

Product intelligence: business

In this section we now address the business drivers and challenges for product intelligence. Apart from its capability to facilitate more robust and flexible operations in manufacturing and supply chain management (Meyer and Wortmann, 2010, Pannequin et al., 2009, Sallez et al., 2009), an intelligent product approach offers an opportunity for the development of more customer-oriented services, such as the potential for controlling of orders during a transportation provider’s operations.

Here, we

Conclusions

This paper has examined 10 years of development in the area of product intelligence (and its related fields). There have been many encouraging initiatives that have helped to articulate the business motivation for such an approach and to identify potentially useful application domains.

Four challenges of particular importance for this subject are:

  • Conceptual: A common agreement on the terminology and definitions associated with product intelligence.

  • Theory: A repeatable methodology for guiding the

Acknowledgments

The authors would like to acknowledge the inputs of colleagues Jin Lung Chirn, Alexandra Brintrup, Tomas Sanchez-Lopez, academic colleagues in the product intelligence field and industrial colleagues at Daimler Chrysler, Boeing, IBM. A preliminary version of this paper was presented to a keynote address at INCOM 2012 in Bucharest, May 2012 (McFarlane, 2012).

Duncan McFarlane is Professor of Industrial Information Engineering at the Cambridge University Engineering Department, and head of the Distributed Information and Automation Laboratory within the Institute for Manufacturing. He has been involved in the design and operation of industrial automation and information systems for twenty five years. His research work is focused in the areas of distributed intelligent automation, product intelligence, reconfigurable industrial systems, RFID

References (101)

  • G.G. Meyer et al.

    Intelligent products: A survey

    Computers in Industry

    (2009)
  • R. Pannequin et al.

    The performance of product-driven manufacturing control: An emulation-based benchmarking study

    Computers in Industry

    (2009)
  • A.K. Parlikad et al.

    RFID-based product information in end-of-life decision making

    Control Engineering Practice

    (2007)
  • Y. Sallez et al.

    A stigmergic approach for dynamic routing of active products in fms

    Computers in Industry

    (2009)
  • A. Sarac et al.

    A literature review on the impact of RFID technologies on supply chain management

    International Journal of Production Economics

    (2010)
  • M. Tajima

    Strategic value of RFID in supply chain management

    Journal of Purchasing and Supply Management

    (2007)
  • S.-F. Tzeng et al.

    Evaluating the business value of rfid: Evidence from five case studies

    International Journal of Production Economics

    (2008)
  • P. Valckenaers et al.

    Intelligent products: Agere versus Essere

    Computers in Industry

    (2009)
  • H. Van Brussel et al.

    Reference architecture for holonic manufacturing systems: PROSA

    Computers in Industry

    (1998)
  • X. Yang et al.

    Intelligent products: From lifecycle data acquisition to enabling product-related services

    Computers in Industry

    (2009)
  • ANSI/ISA (1995). Batch control, part 1: Models and terminology. International Standard ANSI/ISA-S88.01, North Carolina,...
  • U.M. Apte et al.

    Effective cross docking for improving distribution efficiencies

    International Journal of Logistics Research and Applications

    (2000)
  • Borangiu, T., Raileanu, S., Anton, F., Parlea, M., Tahon, C., Berger, T., et al. (2011). Product-driven automation in a...
  • Borangiu, T., Raileanu, S., Trentesaux, D., Berger, T., & Iacob, I. (2012). Distributed manufacturing control with...
  • A. Brintrup et al.

    Will intelligent assets take off? Toward self-serving aircraft

    IEEE Intelligent Systems

    (2011)
  • Brintrup, A., Ranasinghe, D., McFarlane, D., & Parlikad, A. (2008). A review of the intelligent product across the...
  • Chirn, J. -L., & McFarlane, D. (2000). A holonic component-based approach to reconfigurable manufacturing control...
  • N. Chokshi et al.

    A distributed architecture for reconfigurable control of continuous process operations

    Journal of Intelligent Manufacturing

    (2008)
  • comScore (2012). U.S. e-commerce sales growth from 2002 to 2011. Technical...
  • S. Deen

    Agent-based manufacturing: Advances in holonic approach

    (2003)
  • Department of Energy and Climate Change (2012). Annual prices of fuels purchased by manufacturing industry (original...
  • J. Ferber

    Multi-agent systems: An introduction to distributed artificial intelligence

    (1999)
  • FIPA (2002). Standards of the foundation for intelligent physical agents....
  • K. Främling et al.

    Requirements on unique identifiers for managing product lifecycle information: Comparison of alternative approaches

    International Journal of Computer Integrated Manufacturing

    (2007)
  • Främling, K., Holmström, J., Ala-Risku, T., & Kärkkäinen, M. (2003). Product agents for handling information about...
  • Främling, K., Loukkola, J., Nyman, J., & Kaustell, A. (2011). Intelligent products in real-life applications. In...
  • Främling, K., & Nyman, J. (2008). Information architecture for intelligent products in the internet of things. In V....
  • Främling, K., & Nyman, J. (2009). From tracking with RFID to intelligent products. In ETFA 2009: IEEE conference on...
  • Ganji, F., Kluge, E., & Scholz-Reiter, B. (2010). Bringing agents into application: Intelligent products in autonomous...
  • V. Giannikas et al.

    Product intelligence in intermodal transportation: The dynamic routing problem

  • J. Holmström et al.

    Implementing collaboration process networks

    The International Journal of Logistics Management

    (2002)
  • K. Hribernik et al.

    The product avatar as a product-instance-centric information management concept

    International Journal of Product Lifecycle Management

    (2006)
  • Hribernik, K., Warden, T., Thoben, K., & Herzog, O. (2010a). An internet of things for transport logistics – An...
  • Hribernik, K. A., Pille, C., Jeken, O., Thoben, K. -D., Windt, K., & Busse, M. (2010b). Autonomous control of...
  • K.A. Hribernik et al.

    Intelligent products to support closed-loop reverse logistics

  • M. Hülsmann et al.

    Understanding Autonomous Cooperation and Control in logistics: The impact of autonomy on management, information, communication and material flow

    (2007)
  • Jarvis, D., Jarvis, J., McFarlane, D., Lucas, A., & Ronnquist, R. (2001). Implementing a multi-agent systems approach...
  • R. Jedermann et al.

    The benefits of embedded intelligence: Tasks and applications for ubiquitous computing in logistics

  • Kango, R., Moore, P., & Pu, J. (2003). Smart appliances and product service system. In Proceedings of the international...
  • T. Kelepouris et al.

    Bayesian supply chain tracking using serial-level information

    IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews

    (2011)
  • Cited by (0)

    Duncan McFarlane is Professor of Industrial Information Engineering at the Cambridge University Engineering Department, and head of the Distributed Information and Automation Laboratory within the Institute for Manufacturing. He has been involved in the design and operation of industrial automation and information systems for twenty five years. His research work is focused in the areas of distributed intelligent automation, product intelligence, reconfigurable industrial systems, RFID integration, track and trace systems and valuing industrial information. Most recently he has been examining the role of automation and information solutions in supporting industrial services and infrastructure. Professor McFarlane is also Co-Founder and Chairman of RedBite Solutions Ltd – an industrial RFID and track and trace solutions company. He was Professor of Service and Support Engineering from 2006 to 2011 which was supported by both Royal Academy of Engineering and BAE Systems. Since 2010 he has been the Cambridge Professor of Industrial Information Engineering and is Co Investigator in the Cambridge Centre for Smart Infrastructure and Construction.

    Vaggelis Giannikas is currently a PhD student and teaching assistant in the Institute for Manufacturing at the University of Cambridge working on Product Intelligence and distributed decision making. He holds a BSc on Management Science and Technology from the Athens University of Economics and Business (Greece). He is also the Departments chief of XRDS, the ACM magazine for students, and an editor of OR/MS Tomorrow, the INFORMS student magazine. His research interests include industrial informatics, product intelligence, decision support systems and open source software.

    Alex C.Y. Wong is the CEO & Co-Founder of RedBite Solutions. He is an expert in RFID-based supply chain inventory control and has been involved in evaluating the impact of RFID on supply chains since he joined the former Auto-ID Center in 2001. He was an Associate Director of Cambridge Auto-ID Lab and has provided RFID training and consulting services to numerous global organisations. In the past year, Alex was responsible for leading University of Cambridge in an European Union (EU) integrated project called BRIDGE in collaboration with over thirty organisations.

    Mark Harrison is Director of the Auto-ID Lab at the University of Cambridge and has been involved in the development of the EPC Network architecture since 2002. He has been and active participant and co-chair in a numbmer of GS1 standards work groups. Over the last 10 years he has been involved in a number of research projects involving different industry sectors including healthcare, aerospace, consumer electronics. He was also involved in the EU FP6 BRIDGE project (which was co-ordinated by GS1) and involved over 30 participating organizations. One of the main research areas in the Auto-ID Lab at Cambridge is the use of traceability information to make better decisions in industrial operations and logistics processes. Mark has a PhD in physics from the University of Cambridge.

    This article has been written on the basis of the plenary lecture presented at INCOM-12, 14th IFAC Symposium on Information Control Problems in Manufacturing, Bucharest, Romania, May 23–25, 2012.

    View full text