Blog

Integrating technology roadmapping and portfolio management at thefront-end of new product developmentMaicon G. Oliveira ?, Henrique RozenfeldDept. of Production Engineering, São Carlos School of Engineering, University of São Paulo (USP), São Carlos, SP, Brazila r t i c l e i n f o a b s t r a c tArticle history:Received 12 November 2009Received in revised form 9 July 2010Accepted 14 July 2010Many authors point out that the front-end of new product development (NPD) is a criticalsuccess factor in the NPD process and that numerous companies face difficulties in carrying itout appropriately. Therefore, it is important to develop new theories and proposals that supportthe effective implementation of this earliest phase of NPD. This paper presents a new method tosupport the development of front-end activities based on integrating technology roadmapping(TRM) and project portfolio management (PPM). This new method, called the ITP Method, wasimplemented at a small Brazilian high-tech company in the nanotechnology industry to explorethe integration proposal. The case study demonstrated that the ITP Method provides asystematic procedure for the fuzzy front-end and integrates innovation perspectives into asingle roadmap, which allows for a better alignment of business efforts and communication ofproduct innovation goals. Furthermore, the results indicated that the method may also improvequality, functional integration and strategy alignment.© 2010 Elsevier Inc. All rights reserved.Keywords:Front-endNew product developmentTechnology roadmappingProject portfolio managementHigh-tech companiesBrazilian case study1. IntroductionThe front-end of new product development (NPD) comprises the activities that precede the formal development of newproduct projects. This phase defines the new products that should provide competitiveness and revenue for the business, whichmakes it a critical phase for NPD process performance. If a company is ineffective at this point, despite its successful technicaldevelopment, there is a high probability of product failure in terms of financial, strategic or commercial expectations [1–9].Research on NPD process performance has shown the need for improving this initial phase [10–13]. Front-end inefficacies arerelated to features such as cross-functional uncertainties, multiple goals, lack of information and dynamic decisions [2–4, 8].Nevertheless, companies should learn how to deal with the front-end to gain a competitive advantage and to launch successfulproducts.Among many management practices currently applied at the front-end, technology roadmapping and portfolio managementhave been widely adopted because of their capability of achieving effective outcomes [9, 14].At the front-end, technology roadmapping can be used mainly for strategic planning, product planning, program planning andintegration planning [15]. Some advantages of applying TRM are integrating innovation perspectives (market, product andtechnology), facilitation of intraorganizational communication and long-term planning [16].Project portfolio management (PPM) supports the evaluation, selection, prioritization and control of the company’s projectportfolio. Consequently, it helps to ensure strategic alignment, maximization of project portfolio value and resource planning [17],facilitating the selection of the best new product concepts for the development and launch of successful products. This is one of themain front-end objectives.Technological Forecasting & Social Change 77 (2010) 1339–1354? Corresponding author.E-mail addresses: [email protected] (M.G. Oliveira), [email protected] (H. Rozenfeld).0040-1625/$ – see front matter © 2010 Elsevier Inc. All rights reserved.doi:10.1016/j.techfore.2010.07.015Contents lists available at ScienceDirectTechnological Forecasting & Social ChangeSince TRM and PPM have complementary features, some authors have indicated their integration as an opportunity to improvethe front-end [14, 18]. This makes sense, because on the one hand the standalone application of TRM does not result in a portfolioof NPD projects and on the other hand PPM needs a list of projects as input which should be future-oriented or aligned with thecompany’s strategy. Therefore, if PPM is applied without this orientation, i.e., as a standalone tool, it may have a limited listconcerning future orientation. Consequently, the results of the front-end based on the standalone application of TRM or PPM maybe ineffective for the NPD process.Nevertheless, there is little evidence of their integration. In our review of the literature we found only two studies focusing on thistopic: Park and Park [19] and Albright and Nelson [20]. Moreover, both propose roadmapping as an alternative to traditional portfoliomanagement practices, i.e., for selecting and planning a portfolio of projects, rather than focusing on its complementary application.In this context, this research intends to explore the complementary features of TRM and PPM at the front-end of NPD. Toachieve its aim, this paper proposes a method based on the integration of TRM and PPM. The acronym of the proposed method isthe ITP Method, which stands for (I) Integrated, (T) Technology Roadmapping and (P) Portfolio Management. Using a caseresearch methodology, this integrated method was implemented at a small Brazilian high-tech company in the nanotechnologyindustry with the purpose of testing the integration itself and of analyzing benefits for front-end execution. Thus, it contributes notonly to the theory by analyzing how TRM and PPM would be integrated, but also to practice by developing an initial reference forcompanies interested in improving their application of TRM and PPM.The next section reviews the literature of the front-end of NPD, TRM and PPM to contextualize the research and clarify itstheoretical backgrounds. Section 3 describes the methodology adopted in the research. Section 4 then presents the developmentand proposal of the ITP Method. Section 5 describes the case study and the analysis of integrated application of TRM and PPM.Finally, Section 6 concludes with a description of how the integrated application worked during the case study, giving insightsregarding its contribution to front-end execution and suggesting opportunities for future research.2. Literature review2.1. The front-end of new product development (NPD)The front-end of new product development has caught the attention of researchers and companies. Some of the reasons for thisinterest stem from its strong impact on the success of NPD and its ineffectiveness when compared to other parts of the NPD process[2, 4, 10, 11, 13]. Moreover, the front-end creates a link between business goals and the new product development process [6, 21],which makes it an important channel to achieve successful innovations.The activities of the front-end of NPD precede product design, determining product opportunities in terms of strategic goals,market needs and technological solutions. These, in turn, lead to product concepts and to projects to deliver these concepts. Uponconclusion of these activities, the pool of product projects is evaluated, and the ones that converge with business interests areselected for development in the design phase of NPD [3–5].The front-end groups people from many functional areas of business, such as: senior management, marketing, design andmanufacturing [22]. In addition, it works with complex and uncertain information [1, 3, 4]. Other characteristics that challengecompanies during front-end execution are the application of tools from different knowledge areas and the difficulty in definingresponsibilities among stakeholders, which has an impact on its management [23, 24].2.2. Technology roadmappingTechnology roadmapping serves to describe the market, to plan product and process development, to establish technologicalcapacities and to analyze resources [25]. It reveals the interrelations among market, product and technology parameters, andidentifies objectives that justify company efforts [26].The roadmap structure should be linked with technology roadmapping objectives in order to achieve the expected results. Themost generic roadmap has a horizontal timeline and three layers: market, product and technology [16]. Fig. 1 depicts a genericroadmap.Fig. 1. Generic roadmap [16].1340 M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–1354The literature describes various characteristics of TRM. A compilation was made in this study, as follows:• The success of technology roadmapping depends on support from an influential stakeholder [15, 27, 28].• The participation of specialists in the process increases the probability of success [15, 27, 29].• Roadmapping is a multifunctional process; thus, the involvement of participants from different functional areas is essential [20,27].• The symbols employed in the roadmap should be known to enable organizational communication [20].• Customization of roadmapping contributes to the effective application of TRM, since it can be used in different environments andfor distinct purposes [15, 30].• The process is considered responsible for adding value in the TRM application [15, 25, 30].2.3. Project portfolio managementProject portfolio management aligns the management of individual projects with the management of business operations [31].It can be defined as a dynamic process in which the portfolio of active projects is reviewed and updated periodically. During theprocess, current projects are reviewed and cancelled and new projects are assessed, selected and prioritized [18].Project portfolio management has particular features, namely, the presence of unknown information, multiple goals andinterrelated projects [17]. More than simply individual project management, it involves the management of a project group toachieve specific business goals [32].Cooper, Edgett and Kleinschmidt [17] explain that PPM should attain four goals: maximization of portfolio value, portfoliobalancing, strategic alignment and resource allocation. The simultaneous achievement of these goals results in better projectportfolios. However, there is a constant conflict among these goals that limits the achievement of an optimum portfolio.3. Research methodologyThe methodological approach adopted here derives from Karl Popper’s [33] logic of scientific discovery, which states that newideas can be proposed as a hypothesis by means of logical deduction and that they must then be tested by empirical applications.This study does not attempt to prove the integration of TRM and PPM as a generic truth. Instead, it attempts to negate it througha process Popper calls falsification. If the method proves empirically useful to a specific case, this means it is temporarily useful tothis kind of organization until it is tested again in another application. If the proposed method withstands detailed and severe tests,it is corroborated.Thus, the research methodology is a combination of two methods: the deductive approach to define the method (hypothesis),and a case research to test it. The methodology is divided into five stages: analysis of front-end models, analysis of TRM processes,analysis of PPM processes, development of the ITP Method, and testing using a case study. These stages are represented in Fig. 2.The analysis of front-end models involves investigating NPD process models focusing solely on the front-end phase. In addition,some models of the front-end of the innovation process are also considered, since some of them involve the front-end of NPD. Thegroup of selected models is further examined to identify the main activities of the front-end and its function in the NPD process.The analysis of TRM and PPM processes is similar. These two stages select the process models for investigating the activities,organizational aspects and tools used in the application of TRM and PPM.The development of the ITP Method comprises three steps: evaluation of the method’s completeness after integration of TRMand PPM, definition of the method’s activities, and definition of the method’s supporting tools. The first step evaluates how theintegration of TRM and PPM supports the front-end of NPD. The second step makes a comparative analysis of front-end, TRM andPPM to define the method’s activities. The third step selects the tools to be used and customizes them to adapt the procedure forapplication.Testing of the model using the case study is based on the case research methodology proposed by Voss, Tsikriktsis and Frohlich[34] to guide the researcher’s activities and approach throughout the practical study. At this stage, this study aims to assess howTRM and PPM work when applied in an integrated way and identify the benefits this integration would bring to the front-endexecution.Fig. 2. Stages of research methodology.M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–1354 1341The case study results are collected from two interviews, the first prior to the application of the method, when the company’sinitial situation in the front-end is evaluated. In the second interview, held at the end of the case study, the team is asked to explainhow the ITP Method supports the front-end in terms of changes in the metrics: quality, functional integration, strategy alignment,duration and costs.4. Integrated technology roadmapping and project portfolio management method (ITP Method)Subsection 4.1 comprises stages 1, 2 and 3 of the methodology. The development of the method is presented in subsection 4.2,which corresponds to stage 4 (Fig. 2). The outcome of this stage is the ITP Method described in subsection 4.3.4.1. Analysis of references to be integrated into the ITP Method4.1.1. Front-end modelsThis study analyzes front-end models with the purpose of selecting one to be used as a reference in the development of the ITPMethod. Four models were considered: Clark and Wheelwright [6], Cooper [4], Koen et al. [3], and Crawford and Di Benedetto [5].These are among the most cited front-end models in the literature of production innovation management, which gives them thedegree of acceptance required for use in this research.To enable the method to be developed, the model to be selected should be comprehensive regarding the components of thefront-end of NPD. Thus, the adopted criteria analyze if the model encompasses the three main components of the front-end ofNPD: product opportunity identification, concept generation, and project selection. Moreover, the level of detail used in thedescription and the presentation of tools are also considered positive factors for the selection, because they allow for a betterunderstanding and implementation of the proposal.The Clark and Wheelwright [6] model shows four front-end activities: Technology Assessment and Forecasting, MarketAssessment and Forecasting, Development of Goals and Objectives, and the Aggregate Project Plan. The first three activities involvethe identification of product opportunities at the front-end; therefore, the authors completely support this requirement. However,the concept generation and project selection components are only partially fulfilled by the Aggregated Project Plan. This modelpresents its activities in detail and includes some application tools.The Cooper [4] model considers a front-end comprising three activities: Discovery, Scoping and Build Business Case; and threegates: Idea Screen, Second Screen and Go To Development. The identification of product opportunities is supported by theDiscovery activity, the concept generation by Scoping and Build Business Case, and lastly, project selection is provided by theGates. Hence, this model is a good front-end reference model. The only characteristic that makes this model less advantageous isthe low level of details in some activities and the lack of tools and examples, which would better support its application.The model proposed by Koen et al. [3] suggests five front-end activities: Opportunity Identification, Opportunity Analysis,Idea Generation and Enrichment, Idea Selection and Concept Definition. Furthermore, this model adopts a new organizationalstructure among its activities, indicating that the front-end is a flexible process. The first and second activities support productopportunity identification and t

he next three represent the concept generation. The last component, which is the projectselection, is not clearly presented in this proposal. Nevertheless, it is considered during the execution of idea selection andconcept definition activities, but it is not described sufficiently to support front-end development. Regarding its explanation, somedetailed information and examples of tools are missing from this model, which may make it difficult to use as a reference forimplementation.The Crawford and Di Benedetto model [5] presents three front-end activities: Opportunity Identification and Selection, ConceptGeneration and Concept/Project Evaluation. In addition, this model places three Gates after each of its activities: Direction, InitialReview and Full Screen. This model comprises one activity for each component of the front-end, i.e., Opportunity Identification andSelection for product opportunity identification, Concept Generation for concept generation and Concept/Project Evaluation forproject selection. Additionally, this model includes tools and examples of how to implement activities.Table 1 presents a compilation of the analysis and indicates that Crawford and Di Benedetto’s model [5] should be the oneselected, because it is more comprehensive than the others regarding the main components of the front-end. In addition, thismodel offers a more detailed explanation, which is important for front-end implementation.It should be pointed out that the scope of this paper does not include a definition of a “new” model for the front-end. Ourintention is to analyze the integration of TRM and PPM, so that other models could also be defined to test the integration. However,in this work, the more comprehensive model serves as a reference for stage 4.Table 1Analysis of front-end models.Models Product opportunity Concept generation Project selection Level of details Clark and Wheelwright [6]SupportedPartially supportedPartially supportedHighCooper [4]SupportedSupportedSupportedLowKoen et al. [3]SupportedSupportedPartially supportedLowCrawford and Di Benedetto [5]SupportedSupportedSupportedHigh 1342 M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–13544.1.2. Technology roadmapping processesThe research aimed to select a process that would allow for the identification of activities and tools used in the application oftechnology roadmapping. Therefore, the criteria for selecting the TRM process were the presence of a well structured set ofactivities for TRM development and the presentation of tools. This study assessed three technology roadmapping processes toidentify one that could be used in the development of the method: Willyard and McClees [25], Albright and Kappel [29] and Phaal,Farrukh and Probert [16]. These processes are among the most cited references in the literature on roadmapping, which supportstheir inclusion in this study.The Willyard and McClees [25] model comprises eight activities for technology roadmapping: Description of Business;Technology Forecast; Technology Roadmap Matrix; Quality; Allocation of Resources; Patent Portfolio; Product Descriptions, StatusReport and Summary Charts; and Minority Report. This process presents the main activities related to TRM development.However, it is unclear regarding the sequence of application or even the relationship between them, which would add complexityto its implementation. Furthermore, it presents examples of tools used in the activities, but does not explain how to use them.The Albright and Kappel [29] model proposes the TRM development in four sections: Market, Product, Technology andSummary/Action Plan. This process describes a set of activities which deal with each issue of TRM in a clear sequence fordevelopment. The only point lacking in this process is a more detailed explanation of how to perform the activities. In addition, theprocess includes tools related to its activities; however, like the previous process, it does not explain how to apply them.The Phaal, Farrukh and Probert [16] model proposes four workshops for TRM: Market, Product, Technology and Charting. Inaddition to the workshops, the process proposes a planning activity at the beginning and an implementation activity at the end. Allworkshops and activities are interlinked using their results, adding a clear flow in the process. Finally, this model offers anextensive explanation concerning the development of each activity and a group of tools for use during this application.The results of the analysis are presented in Table 2. This table indicates that the Phaal, Farrukh and Probert [16] TRM processwas the most suitable one according to the aforementioned criteria, as it contains a detailed and structured procedure oftechnology roadmapping application in terms of activities and tools.4.1.3. Project portfolio management processesNumerous portfolio management processes are available, since this issue has been studied for many years. This researchintends to follow a procedure for portfolio management that represents the standard approach reported in the literature.Consequently, it is possible to increase the generalization achieved by its integration with TRM.Based on this purpose, this study selected the proposals presented by Cooper, Edgett and Kleinschmidt [18] and Archer andGhasemzadeh [35]. These processes are two traditional and accepted approaches for portfolio management that include theactivities and tools required to develop the method, according to a bibliographical review of the portfolio management field.The Cooper, Edgett and Kleinschmidt [18] process involves periodic meetings for portfolio review, during which new andongoing projects are evaluated to define the project portfolio. Additionally, this process proposes gates between project phases.The gates evaluate each ongoing project individually to decide if it should be continued, cancelled or reviewed.The Archer and Ghasemzadeh [35] PPM process comprises three phases: pre-process, process and post-process. The preprocess phase involves definition of the strategy and selection of the process methodology. The process phase comprises thefollowing stages: pre-screening, individual project analysis, screening, project portfolio selection, and portfolio adjustment. Thethird phase, post-process, involves the control of ongoing projects.An analysis of these two methods indicates the presence of a common set of activities, although they include somecomplementary aspects. Archer and Ghasemzadeh [35] propose an in-depth analysis of each project, and Cooper, Edgett andKleinschmidt [18] analyze the balance of the portfolio and resource capacity limitations. Consequently, this study intends tocombine these complementary parts to reach a portfolio management process which would best support integration withtechnology roadmapping.4.1.4. Implications of the selected process for the ITP Method’s featuresListed below are some considerations about the selected references that affect the application and configuration of TRM andPPM in this study:• Since Crawford and Di Benedetto’s [5] model is not designed to support technology planning, the ITP Method considers onlyproduct planning. Moreover, as this is a reference model, the ITP Method must be customized to meet the requirements of aspecific practical application.• Like the ITP Method proposed here, the Phaal, Farrukh and Probert [16] process focuses on product planning. However, the TRMprocess considers technology forecasting as an input for product planning. Since Crawford and Di Benedetto’s [5] model does notTable 2Analysis of TRM processes.TRM processes Organization Tools Willyard and McClees [25]Albright and Kappel [29]Phaal, Farrukh and Probert [16]UnclearClearClearIncluded/not describedIncluded/not describedIncluded/described M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–1354 1343include technology planning, the application of the TRM process adopted in the ITP Method may lack the input of technologyforecasting.• The PPM process proposals of Cooper, Edgett and Kleinschmidt [18] and Archer and Ghasemzadeh [35] were combined in the ITPMethod for the specific application described in this paper. This may limit the proposal, because it was not validated in othersituations. Therefore, an evaluation must be made of the activities that make sense in other specific applications.• The configuration of technology roadmapping and portfolio management is flexible for each context. Due to requirementsintroduced by the front-end of NPD, and the characteristics of TRM and PPM processes adopted, the overall configuration definesTRM as being responsible for product opportunity identification and for confirmation of strategic alignment of the selected NPDprojects. On the other hand, the PPM process will be assigned to evaluate, prioritize and select projects. Thus, following Crawfordand Di Benedetto’s [5] front-end model, the TRM is initially applied to identify product opportunities, after which the PPM isimplemented to evaluate and select projects, and lastly, the portfolio of selected NPD projects is organized into a small roadmap,which is compared to the original one to ensure strategic alignment.4.2. Development of the ITP Method4.2.1. Evaluation of the method’s completeness after integration of TRM and PPMThis step analyzed how technology roadmapping and project portfolio management fit the front-end. The contents and goals ofCrawford and Di Benedetto’s [5] front-end model were compared with Phaal, Farrukh and Probert’s [16] TRM process, and withCooper, Edgett and Kleinschmidt’s [18] and Archer and Ghasemzadeh’s [35] PPM processes.The technology roadmapping process revealed a similarity with the opportunity analysis and selection phase of the NPD frontend model, while the project portfolio management process remained aligned with the concept/project evaluation phase. Therelationship of TRM and PPM with opportunity analysis and selection phase and with concept/project evaluation phase,respectively, contributes two insights:• If these phases follow a sequence in the front-end, the application of TRM and PPM should follow the sequence of phases they arerelated to. Hence, TRM should be introduced before PPM, because the opportunity analysis and selection phase precede theconcept/project evaluation phase of Crawford and Di Benedetto’s [5] model.• Neither technology roadmapping nor project portfolio management contributes consistently to the concept generation phase.Hence, the integration of these methods may not fully meet the requirements to support the front-end of NPD.These insights will affect the definition of method’s activities.4.2.2. Definition of the method’s activitiesThe method’s activities were defined based on the aforementioned models. The activities selected from TRM were: definition ofunit of analysis, business strategy analysis, market analysis, product analysis, technology analysis, and definition of productstrategies. The project portfolio management yielded: financial evaluation, evaluation of probability of success, evaluation ofstrategic alignment, project prioritization, project interrelationship analysis, and selection of the NPD project portfolio.An evaluation was made to determine if the activities defined up to this point from TRM and PPM encompassed thefunctionality expected for the front-end. To this end, the aforementioned activities were compared with the functionalitiesdescribed in Crawford and Di Benedetto’s [5] model, to ascertain if they sufficed to support the proposed method.This comparison indicated that the method lacked an activity related to the concept generation phase. Therefore, a new activitywas included, the “Proposition of NPD projects”, to complete the other activities derived from technology roadmapping andproject portfolio management.In the end, the flow of method activities started with TRM activities, followed by the “Proposition of NPD projects” activity, andending with PPM activities.4.2.3. Definition of the method’s supporting toolsThe last step involves defining the tools that should be used in the application of the method. Most of the necessary tools werefound in the following references: Crawford and Di Benedetto [5], Phaal, Farrukh and Probert [16], Cooper, Edgett andKleinschmidt [18] and Archer and Ghasemzadeh [35]. Complementary references were examined to define tools for financialevaluation [36–38], probability of success evaluation [39] and project interrelationship analysis [32].4.3. ITP Method overviewThe procedure developed for the ITP Method comprises 13 activities which are described in the following discussion. Fig. 3illustrates the method, showing the activities, information flow and the connections among activities. Further information aboutthe method is given in the subsection of each activity.4.3.1. Definition of unit of analysisThis activity establishes the boundaries of the method’s application in terms of business unit, market segment and product line.The unit of analysis should reflect the scope the company wishes to consider in the front-end of NPD. Fig. 4 presents thecomponents of this unit.1344 M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–13544.3.2. Business strategy analysisThis activity identifies the business drivers that establish the directions and targets in the NPD process. First, the methodexamines the performance dimensions of the product line, which represent quantitative and qualitative product aspects such as:weight, originality, innovation, profitability, etc. Based on these dimensions, the team analyzes the business strategies and definesthe drivers related to them. Some examples of business drivers are: growth of market share, improvement of organizationknowledge, increase in business profitability, etc. During the strategy analysis, the start and end dates of business drivers should beidentified to determine their position in time.Finally, the team selected to participate in the activities evaluates the priorities of the business drivers. If the unit of analysisconsiders more than one business unit, the priorities may be different due to the possible existence of independent strategies ineach unit.4.3.3. Market analysisThis activity involves defining the drivers that represent market needs and trends. Like in the previous activity, an analysis ofthe product performance dimensions may help in the identification of market drivers. Some examples of market drivers are:customized products, cheaper products, smaller sizes, etc. The duration and priority of each driver are estimated based on marketrelated information. If the unit of analysis encompasses more than one market segment, the priorities may also differ.Fig. 3. ITP Method overview.Fig. 4. Components of unit of analysis.Adapted from [16].M.G. Oliveira, H. Rozenfeld / Technological Forecasting & Social Change 77 (2010) 1339–1354 13454.3.4. Product analysisThis activity determines the product features of the selected product line. Some examples of product features are type ofmaterial, modular framework, size specification, etc. In order to determine the most important features, an evaluation is made oftheir ability to fulfill business goals and market drivers. Since the drivers may have different priorities, this evaluation alsoindicates possible arrangements that can contribute toward product strategies. For instance, a new product with specific featuresmay be chosen to meet the priorities of a single market segment, or features that are always important to specific market segmentsmay serve as the basis for the creation of a new product platform. To this end, the products are planned to meet the requirementsof their respective drivers in the appropriate time.4.3.5. Technology analysisThis activity involves identifying the technologies and evaluating their ability to deliver the product features. Based on thisinformation, the company can ascertain whether or not the required technologies are available. If unavailable, plans should be madefor their development or acquisition. In so doing, the company ensures they will be ready for use in new product development.4.3.6. Definition of product strategiesThis activity analyzes the information created in the preceding activities to formu