To overcome the issues mentioned in Section 2.1, our interdisciplinary approach is based on the following concepts:

To regard long-term future research for sustainability based on rigorous science and engineering as the basis of RDMOET;

In this paper, “future research for sustainability” is defined as any exploration of what may happen and what may be desirable in the field of sustainability. As mentioned in Subsection 2.1.1, sustainability is a broad concept encompassing the search for balance among environmental, economic, and societal challenges and opportunities [11]. Although the environmental aspect is the principal focus of sustainability research, future societal and economic changes, such as sustainable consumption, dematerialization and de-growth, are also considered in RDMOET for eco-innovation, because innovation is achieved not only by technological breakthroughs but also by changes in the way technology is used in society. Unfeasible future targets and scenarios devoid of engineering rigor mislead R&D. In this paper, it is considered that feasible long-term targets and scenarios grounded in sustainability research should be established based on rigorous science and engineering.

An essential driving force of R&D is the motivation of individual researchers, such as the desire to apply a new principle to a new device. Many researchers are inclined to emphasize the functionality of their R&D theme but not the risks involved. Generally, they are not motivated to evaluate their themes from environmental perspectives. Moreover, because researchers may not recognize their R&D technologies are eco-technologies, the opportunity to identify application fields may be missed. They tend to presume their work is superior to other work in the same research field; moreover, sometimes they are dismissive of other work. Thus, it is useful to evaluate their research work objectively. Clearly, since overmanagement or excessive interposition in research might disturb innovation, research themes should be managed carefully. This means RDMOET should be consistent with the existing R&D management methods, such as 3rd-generation R&D management [15], portfolio management [16] and stage gate [17].

Figure 2 shows a framework of RDMOET consisting of four major processes that are not necessarily executed sequentially and strictly but rather simultaneously and flexibly. Processes (a) and (d) are unique to RDMOET, whereas methods utilized in processes (b) and (c) are similar to those of conventional 3rd-generation R&D management. The processes are as follows:

Future research for sustainability: The crucial basis of RDMOET is future research based on rigorous science and engineering. In this process, forecasting and backcasting scenarios are examined for a broad spectrum of energy, natural resource, social, economic, and other environmental problems. The major outputs of the process are an R&D vision and feasible scenarios for realizing environmental sustainability;

Process (d) in Figure 2 is shown in detail in Figure 3.

The processes shown in Figure 3 are explained as follows: (d1)

Evaluation of all R&D themes using a checklist: In Step 1, a checklist approach is applied to all R&D themes at regular intervals. A checklist should be easy for researchers to use. An eco-value checklist of R&D themes was developed by researchers. This checklist approach is based on benchmarking against the current best practice, and the checklist covers energy saving, low carbon emission, material saving, reuse/recycle, chemical risk reduction, etc. The indirect effects of the target technology are assessed for each viewpoint explicitly. The evaluation of indirect effects is useful for enhancing the positive effects of technology;

In this subsection, an example of futures research concerning electricity demand and supply is explained. With regard to climate change, which is the foremost global environmental issue, long-term targets for reducing greenhouse gas (GHG) emissions have been discussed [8]. Since GHG reduction targets strongly depend on characteristics of countries, such as their policies and the nature of their economies, they are not necessarily adequate for other countries' situations. A scientific approach is necessary. Komiyama estimated the future target for sustainable technology development in Vision 2050 based on three preconditions [12], namely:

Developing countries are guaranteed the right to modernize;

As an estimation result, he concluded there should be three main targets of Vision 2050 as follows:

A three-fold increase in energy use efficiency;

In that estimation, however, the detail of the calculation is not clarified. For example, considerations of electrification and progress of ICT are inadequate. Moreover, the uncertainty of some important parameters, such as installation of non-fossil fuel power and energy intensity of developing countries, is not considered. We have tried to estimate the target values of energy-use efficiency and low-carbon power generation for halving GHG emissions compared with the current level. The preconditions of our estimation are as follows:

Social and economic:

The estimation of population growth is assumed. Namely, the world's population will be 9.2 billion in 2050;

First, we analyzed energy consumption per capita in the countries and regions of the world, and estimated the effect of electrification (Figure 4). In the analysis and estimation, progress of ICT [34] and countermeasures for water stress in developing countries [35] were taken into account.

In the IEA Blue Map scenario [7], it is estimated that global electricity generation will be 42,340TWh in 2050. In the Blue Map scenario, the ratio of electricity generated by non-fossil fuel in 2050 is assumed to be 70% of the total supply, whereas it was 33% in 2005. Figure 5 illustrates the configuration of electricity generation in the Blue Map scenario.

Next, the estimated electricity consumption per capita is shown in Figure 6. Here, the “2050 Sustainable Electrification” scenario is derived to satisfy the previous preconditions. This means electricity consumption must be reduced by 60% compared with the 2050 Electrification BAU scenario for GHG stabilization. In other words, a two-and-a-half-fold increase in the efficiency of electricity use is necessary by 2050. Because loss of electricity during transmission and distribution (T&D) is ignored in the estimation, we have to consider T&D loss. As a result, almost the same target of energy saving is derived as in Vision 2050 [12] from slightly different preconditions.

Finally, uncertainty is considered (Figure 7). If the ratio of electricity generated with non-fossil fuel is 60%, the target value of energy saving increases to 90%. Also, modernization in developing countries, namely social and economic precondition 2, affects the target area. In any event, the estimation result shows that a challenging target is necessary for both the demand-side energy saving and the non-fossil fuel electricity generation.

In addition to the issue of climate change mitigation, some important sustainability issues are studied, such as mineral resource supply, chemical hazard and biodiversity, as themes for future research. We believe that doing such future research and sharing the results will be useful, because we can attain a relatively realistic grasp of the future situation when developing ideas for eco-themes.

We assume that the difference of environmental performance could be converted to a monetary value through competition. Since it is difficult to estimate product prices derived from the target technology in the R&D process, we focus on environmental performance. A basic concept is not complete LCA [28] but benchmarking using adequate metrics, such as a streamlining approach [30], indicator for material saving TMR [31], indicator for toxicity TPI [33] and evaluation of ICT solutions [29]. At the Corporate R&D Center of Toshiba Corporation, R&D activities cover a wide spectrum from nanotechnology to ICT, and from basic to applied research. Thus, the evaluation methodology has to cover these various fields.

We consider 3rd-generation R&D management [15] and operational procedures of an enterprise's EMS [26] to be important reference management theories comparable with RDMOET, given that 3rd-generation R&D management remains a theory widely applied in industry and EMS is the basis of the environmental activities of enterprises. In general, the two management systems are independent of each other. One major reason is that an EMS process is virtually routine work as a PDCA cycle, whereas an R&D management process is agile because it is linked to changes in the business environment. Therefore, it is difficult to integrate these two management processes. Another reason is that the divisions responsible for the two management processes are different. For example, R&D management is usually promoted by the research planning division, whereas EMS is usually promoted by the environmental division.

In contrast, RDMOET can link 3rd-generation R&D management and EMS. The eco-value evaluation process of RDMOET is regarded as periodic, virtually routine work (Figure 3), and this evaluation process can be managed as a PDCA cycle of EMS. For instance, planning and execution of R&D themes is done by the research division, the check process using eco-value evaluation is promoted by eco-technology management, and action is taken by the R&D board. Furthermore, processes (b) and (c) in Figure 2 are agile and synchronized with 3rd-generation R&D management. As a result, the eco-theme portfolio and roadmap reflecting the eco-value evaluation results are always shared by the R&D board, and eco-technology is recognized in R&D management. However, decision-making cannot be executed systematically when a conflict arises between long-term and short-term requirements. In such a case, the enterprise's philosophy or policy is reflected in the decision.

In 3rd-generation R&D management, the focus is on actualized market needs because R&D is tightly linked to a business plan, whereas consideration of potential needs is weak. The EMS procedure also cannot treat potential environmental needs, only actualized environmental issues. In contrast, process (a) in Figure 2 plays an important role in the discovery of potential needs for eco-innovation as a basis of RDMOET.

In this sub-section, the advantages, the disadvantages, the lessons and the limitations of the proposed framework are summarized.

There are three advantages. The 1st advantage is that it provides a systematic way to bridge between long-term environmental sustainability and medium- or short-term research work plans. Since the proposed framework is familiar to those who have experience of 3rd-generation R&D management and EMS, it is easy to put it into practice.

The 2nd advantage is that the RDMOET framework realizes data-driven management of eco-technology through accumulation of eco-value evaluation results. Eco-value of research themes can be categorized into patterns in research fields. These eco-value patterns help overcome the valley of death that must be traversed in the transition from R&D to the market by viewing a research theme from a different perspective, such as in terms of electricity saving or less material use. This is an organizational learning effect at the management level.

The 3rd advantage is the change in the mindset of individual researchers brought about by their evaluation of their own research themes, with technical support. We have observed the increase of environmental awareness in both R&D management and among researchers brought about by the proposed approach. In other words, we recognize that the eco-value evaluation promotes organizational learning that contributes to evolution of the R&D process. However, it is difficult to quantify the effect of the organizational learning. That is a subject for future work.

The disadvantage of the framework is there is a risk of losing the diversity of original ideas if operation of the framework is excessively strictly. For instance, even though a negative result is derived from an eco-value evaluation, care must be exercised so as not to reduce researchers' motivation. With regard to application and operation of the framework, one must proceed with care. An important lesson from our practice is that abolishing research themes should be avoided to ensure sufficient diversity of research themes if possible. Rather, focusing on monitoring and correcting the direction of research themes with potential eco-value drives change of a research organization toward more active eco-innovation.

Our framework has three limitations. The 1st limitation is that it is not general but specifically designed for the Corporate R&D Center of Toshiba Corporation. Therefore, the methodology presented here should be modified if it is applied to other research institutes. For example, the proposed RDMOET framework has to be modified for an open innovation environment such as the 4th-generation R&D management framework. R&D management is also highly dependent on know-how and organizational culture. Although their explicit representation is important for improving the proposed framework by making it more generally applicable, it is inherently difficult to share practice examples and organizational cultures.

The 2nd limitation is that the proposed framework cannot be executed without managerial vision and strong leadership to promote RDMOET. For instance, the CEO's commitment is a powerful factor and Environmental Vision 2050 is shared throughout Toshiba Group [39], providing a managerial foundation for promoting RDMOET. Additionally, RDMOET is promoted throughout the Corporate R&D Center of Toshiba Corporation by a department established for that purpose. The department is the linchpin of RDMOET operation and its existence attests to a powerful commitment.

The 3rd limitation concerns future research and eco-value evaluation. Future research methodology involves many quantitative/qualitative, normative/exploratory models and methods [6]. Since a single research organization cannot conduct future research covering all fields, a methodology is required in order to compare or integrate various future scenarios [40,41]. For evaluation of new technology, it has been pointed out that eco-efficiency at the micro level is not necessarily consistent with sustainability at the macro level [42]. Therefore, it is important to pay attention to the estimation of the macro effect of the target eco-technology. For instance, the rebound effect, which is one of the side effects of new technology, has recently become a focus of interest [43], and we have developed an indirect rebound effect estimation method [44]. It is necessary to systematize the relationship between evaluations at the micro and macro levels theoretically.