Nonassembled products

As a first approximation, we can sort the universe of goods produced into two categories. The first is assembled products, manufactured goods that are built from discrete parts. Automobiles, fountain pens, and personal computers are among these, and their production is measured in units. In contrast, nonassembled products often become part of assembled products— the paint on the car, the glass on the windshield, the gasoline in the tank, and so forth. Nonassembled products can often be expressed as a chemical formula and are usually measured in terms of weight or volume rather than units. 

Innovation comes to both assembled and nonassembled products in one of two ways. The innovation can be radical, offering a major departure from current practice and a discontinuous leap in some aspect of performance. Or it can be incremental, a series of small improvements, often undocumented, that accumulate to large gains. Incremental innovation underlies the continuous improvement made famous by Japanese manufacturers or the learning curve well known around the world. 

Hydrogen is an archetypal nonassembled product, and the devices that convert it to services desired by consumers are archetypal assembled products. Thus, hydrogen use in transportation requires innovations at a roughly matching pace on both the top row and the bottom row of the innovation matrix, shown for the hydrogen transition in Fig. 7-2. Stimulating these innovations for the public purpose presents the DOE with a significant... 

Finally, incremental process improvements in the form of continuous, small-scale innovations add significantly to the productivity of mature nonassembled products. There, innovation is almost exclusively the province of the incumbent producers who are motivated chiefly by cost competition. When combined with the leaps forward made by the few radical innovations, productivity gains tend to follow the pattern shown in Fig. 7-3. In glass manufacture and similar industries, the productivity gains that accrue from these incremental changes and those from radical changes in process architecture appear about evenly divided (Utterback, 1994). The hydrogen experience would probably be close to that shown in Fig. 7-3. 

---