Managing platform architectures and manufacturing processes for nonassembled products

The article presents methods for defining product platforms and measuring business performance in process intensive industries. We first show how process intensive product platforms can be defined using the products and processes of a film manufacturer. We then present an empirical method for understanding the dynamics of process intensive platform innovation, allocating engineering and sales data to specific platform and product development efforts within a product family. We applied this method to a major product line of a materials manufacturer. We gathered ten years of engineering and manufacturing cost data and allocated these to successive platforms and products, and then generated R&D performance measures. These data show the dynamic of heavy capital spending relative to product engineering as one might expect in a process intensive industries. The data also show how derivative products can be leveraged from underlying product platforms and processes for nonassembled products. Embedded within these data are strategies for creating reusable subsystems (comprising components, materials, etc.) and common production processes. Hard data on the degree to which subsystems and processes are shared across different products frequently are typically not maintained by corporations for the duration needed to understand the dynamics of evolving product families. For this reason, we developed and applied a second method to assess the degree of reuse of subsystems and processes. This method asks engineering managers to provide subjective ratings on an ordinal scale regarding the use of technology and processes from one product to the next in a cumulative manner. We find that high levels of reuse generally indicate that a product family was developed with a platform discipline. We applied this measure of platform intensity to two product lines of integrated circuits from another large manufacturer. We used this method to gather approximately ten years of information for each product family. Upon analysis, one product family showed substantial platform discipline, emphasizing a common architecture and processes across specific products within the product line. The other product family was developed with significantly less sharing and reuse of architecture, components, and processes. We then found that the platform centric product family outperformed the latter along a number of performance dimensions over the course of the decade under examination.     

Konsonanz 30 Jahre später

Consonant coverage is assumed to be an important prerequisite for media effects to occur. However, empirical evidence for this assumption is largely missing. Focusing on the issue of European integration, this study investigated, in 13 EU member states, whether consonant coverage affected EU citizens’ opinions on the issue. For each country and at the individual level of analysis, the study linked content analyses of the main evening news and the most prestigious newspaper with survey data. Consonant coverage influenced people’s opinions about European integration, while dissonant coverage made no impact. Both EU supporters and EU opponents were susceptible to the effects of consonant coverage, which tentatively suggests that consonant coverage may surmount selective mechanisms. In sum, the study is the first to provide evidence that consonant coverage may indeed enhance media effects.     

Beneficial changes in random variables under multiple sources of risk and their comparative statics

The consequences of a change in a random parameter are determined for a decision model with more than one source of randomness. The two cases of independent and stochastically dependent sources of risk are discussed. Four comparative static theorems are given. These state the effect of first degree stochastically dominant shifts or risk decreases for one random variable while the other random variable is held fixed. Deterministic transformations are used to represent random parameter changes. The results are presented in the context of the coinsurance demand model with a risky insurable asset and background risk.