Information Storage Industry Center

A 3 year study of global employment patterns has been undertaken for three major sectors of the hard disc drive industry. The investigation shows the shift away from high cost centers to mainly Asian based low cost locations and the movement within Asia to find lower overheads or preferred centers of production.

A decade of stagnation and financial crisis have discredited the heavy-handed industrial policies of Japan and Korea, particularly preferential allocation of capital to large companies. In contrast, the remarkable growth of specialized yet flexible computer and semiconductor firms in Taiwan has been supported by a very different type of industrial policy. Taiwan's "ITRI model" combines development initiative and engineering support from the quasi-governmental Industrial Technology Research Institute with commercialization by a mass of small to medium-sized firms in the adjacent Hsinchu Science-Based Industrial Park. The government provides technological support for both long-term development and specific innovative products, but it has exercised extreme restraint in the allocation of capital and sponsored the development of a thriving venture capital industry.

Some industries, however, have proved resistant to the ITRI model. A prime example is hard disk drives. Hard disks combine high barriers to entry and brutally short product cycles. Moreover, they lack the stable architecture underlying notebook computers or monitors. Even with help from ITRI, Taiwan's small firms proved unable to keep up with the rapid development and growing economies of scale in hard disks. However, when they failed at disk drives, they proved capable, with additional assistance from ITRI, of moving quickly to related products with much lower barriers to entry, such as CD-ROMs and CD-R disks and drives.

Not withstanding the limitations revealed by the hard disk case, Taiwan's ITRI model is likely to attract increasing attention from developing countries that recognize the dangers of the Korean approach, yet hope to improve their standing in the international division of labor.

Many high-tech industries are shifting their focus from minimizing time-to-market to minimizing time-to-volume. This puts the tail end of product development, the production ramp-up, in a critical position. This paper presents a case study of product transfer and production ramp-up in the hard disk drive industry. We provide a detailed description of the ramp-up period. By documenting detailed time-series data of several operational measures, we also shed light on the various forces that allow an organization to increase its production volume. Finally, the setting of our research allows us to study product transfer from development in the US to an off-shore production facility.

Rapid product lifecycles and high development costs pressure manufacturing firms to cut not only their development times (time-to-market), but also the time to reach full capacity utilization (time-to-volume). The period between completion of development and full capacity utilization is known as production ramp-up. During that time, the new production process is ill understood, which causes low yields and low production rates. This paper analyzes the interactions among capacity utilization, yields, and process improvement (learning). We model learning in the form of deliberate experiments, which reduce capacity in the short run. This creates a trade-off between experiments and production. High selling prices during ramp-up raise the opportunity cost of experiments, yet early learning is more valuable than later learning. We formalize the resulting intertemporal trade-off between the short-term opportunity cost of capacity and the long term value of learning as a dynamic program. The paper also examines the tradeoff between production speed and yield/quality, where faster production rates lead to more defects. Finally, we show what happens if managers misunderstand the sources of learning.

The hard disk drive industry has been under great cost pressures. Manufacturing has achieved very high levels of efficiencies and there is hardly any room for reducing costs any further by improving manufacturing. An area worth exploring is the design of the hard drives to further reduce the costs. Modular design helps in developing designs that will be amenable to cost reductions by identifying those components that could be designed independently of the rest of the product. In this paper we describe how modular design can accommodate technological innovations. We then relate it to the hard drive industry, and examine how hard disk drives have incorporated the technological innovations. We describe a model to determine a component’s and product’s modularity in a quantitative way. The index developed can then be used to allocate design resources in an efficient way.

Magnetic tape has long been used to backup computer information, customarily during dedicated, low use, 'overnight windows' when tape backup programs could be run without interfering with user applications. Over time these windows have shrunk to near non-existence due to the globalization of business through its use of the Internet and the World Wide Web. While tape system speeds have accelerated and tape capacities have increased, they have not keep pace with the demand for shorter backup windows, quicker restoration requirements, and the rapidly escalating volume of disk drive data being backed up. At the same time, the cost of PC disk drives ("ATA" computer interface) has dropped to be competitive with tape, and disk-to-disk (D2D) backup has become popular, particularly using the new Serial ATA (SATA) PC drives. A D2D system can run at the full speed of disk, and can use the higher capacity of PC disk drives (up to 400 Gbytes today), because backup is primarily serial data storage not needing the high high random access speed of the enterprise storage systems being backed up (which get high random access speed by running many smaller capacity disks in parallel). The current interest in SATA D2D backup is of concern because these are PC drives sold with narrow profit margins that limit drive reliability.

This paper addresses the fact that the reliability of SATA PC drives is inherently lower than enterprise-class drives (SCSI and Fiber Channel "SCSI/FC"). Methods are proposed for SATA storage system designers to achieve high system level reliability by requiring appropriate SATA drive reliability testing, by increased RAID redundancy, and by system management of drive failure warnings. A method is proposed for maintaining user data security in removable D2D archival drives.

This paper investigates how the structure of market demand affects the nature and extent of competition over consumer subgroups in the market. It develops an analytic model to examine how the satisfaction of consumers¹ requirements and the relationships between consumers¹ preferences interact to affect competitive interactions. The model, tested using simulation, reveals demand-side influences on the emergence of three distinct competitive regimes: isolation, in which technologies do not interact throughout the course of their evolution; convergence, in which technologies evolve to compete head-on for the same consumer groups; and displacement, in which one technology cedes dominance of its home market to its rival. The model highlights the critical role played by price in influencing technology displacements and sheds some new light, supported by empirical data from the disk drive industry, on the important phenomenon of disruptive technologies.

When and where will a new organizational form emerge? Recent theory says that as the number of organizations using a particular external identity code first increases beyond a critical minimal level, the code becomes an organizational form. But how is an external identity code established? We assume that the identity code derives from the aggregated identities of individual organizations. Our core argument holds that when the identities of individual organizations are perceptually focused, they will more readily cohere into a distinct collective identity. We develop ideas about how two observable aspects of organizations might generate perceptually focused identities in a common market: (1) de novo entry and (2) agglomeration in a geographic place with a related identity. Using comprehensive data from the market for disk drive arrays, we analyze these ideas and an alternative by estimating effects of different specifications of organizational and product densities on rates of entry and exit for array producers. The findings show that the density of de novo firms affects all (de alio as well as de novo) disk array producers in form-establishing ways: de novo density significantly increases all firm entry and significantly reduces all firm exit. Analyzing densities of certain geographic areas, we also find evidence of faster form development in a place with a related identity and a geographic agglomeration of disk array producers. Finally, we find that joint operation of the two processes, geographic agglomeration of de novo producers in a place with a related identity, serves to enhance form emergence even faster. Overall, the analysis supports the notion that firms with perceptually focused identities aid in establishing an organizational form. It does not show empirical support for a common sense alternative interpretation based on product proliferation.

Making goods evolved over several centuries from craft production to complex and highly automated manufacturing processes. A companion paper by R. Jaikumar documents the transformation of firearms manufacture through six distinct epochs, each accompanied by radical changes in the nature of work. These shifts were enabled by corresponding changes in technological knowledge. This paper models knowledge about manufacturing methods as a directed graph of cause–effect relationships. Increasing knowledge corresponds to more numerous variables (nodes) and relationships (arcs). The more dense the graph, the more variables can be monitored and controlled, with greater precision. This enables higher production speeds, tighter tolerances, and higher quality.

Changes in knowledge from epoch to epoch tend to follow consistent patterns. More is learned about key classes of phenomena, including measurement methods, feedback control methods, and disturbances. As knowledge increases, control becomes more formal, and operator discretion is reduced or shifted to other types of activity. Increasing knowledge and control are two dimensions of a shift from art towards science.

Evolution from art to science is not monotonic. The knowledge graphs of new processes are riddled with holes; dozens of new variables must be identified, understood, and controlled. Frederick Taylor pioneered three key methods of developing causal knowledge in such situations: reductionism, using systems of quantitative equations to express knowledge, and learning by systematic experimentation.

Using causal networks to formally model knowledge appears to also fit other kinds of technology. But even as vital aspects of manufacturing verge on “full science,” other technological activities will remain nearer to art, as for them complete knowledge is unapproachable.