Select an Evaluation Tool

Using a single evaluation tool or methodology throughout the baseline assessment helps ensure consistency among evaluations conducted at different facilities and by different individuals. In addition, remember that you will need to consolidate all the information you gather during the assessment phase, and it s far easier to do so if you have used a common method. 

FIGURE 4-1. Process Safety Management Assessment Framework Source Arthur D. Little, Inc. 

Keep these general principles in mind as you consider the following discussion of PSM assessment tools they can help you determine the best approach for your company. 

In the absence of explicit criteria, the seven characteristics listed below may be useful in evaluating management systems. Keep In mind that these characteristics are not absolute requirements facility management systems may vary significantly and still be capable of achieving the desired results. 

Definition of Responsibilities Facility personnel understand their roles and responsibilities in achieving the desired level of PSM performance. Appropriate checks and balances have been establish to minimize confiicts of interest. Internai coordination and communication mechanisms exist. 

---

At the fundamental level of equilibrium modeling the advantages are many. The model can combine a number of compartments through simple relationship to describe a realistic environment within which chemicals can be ranked and compared. Primary compartments that chemicals will tend to migrate toward or accumulate in can be identified. The arrangement of compartments and their volumes can be selected to address specific environmental scenarios. Data requirements are minimal, if the water solubility and vapor pressure of a chemical are known, other properties can be estimated, and a reasonable estimate of partitioning characteristics can be made. This is an invaluable tool in the early evaluation of chemical, whether the model be applied to projected environmental hazard or evaluation of the behavior of a chemical in an environmental application, as with pesticides. Finally, the approach is mathematically very simple and can be handled on simple computing devices. 

The selection of variables is of central importance for the outcome of a system comparison on environmental and resource use impacts. The ideal variable or set of variables respectively provides information and describes the state of environmental phenomena with certain significance. Thus, applying a set of variables should make it possible to monitor and assess the state of the environment, to identify changes and trends, to transmit scientific data to become relevant for policy, and to evaluate already implemented policy measures. The concept of environmental indicators is broadly accepted as an adequate tool. Accordingly, an indicator is defined as a parameter or a value derived from parameters, which indicates the state of the environment with significance extending beyond that which is directly associated with a parameter value. A parameter s definition in this context is a property that is measured or observed (OECD 1994). Fieri et al. (1996) states that the purposes of indicators are as follows ... 

Informed substitution enables the move from the use of chemicals of concern to safer chemicals, while minimizing the likelihood of unintended consequences. This technique has been demonstrated to be an effective tool and is recognized [8] for guiding the evaluation of chemicals or products in the context of functional need. The options for meeting a functional need can be arrayed along a spectrum of preference based on environmental and human health characteristics. Decisions based on informed substitution will avoid an inclination to select an alternative that may appear more benign only because it is poorly characterized. 

With the introduction of more economical, large-format (320 x 256 pixels) FPA detectors, NIR hyperspectral imaging was also developed into an important tool in selected industrial areas such as the evaluation of food quality (see Chapter 8) or in the pharmaceutical industry. Most likely, the delay in implementation of NIR... 

The solution presented here is based on CORBA [877], a wide-spread and mature middleware platform. CORBA provides direct support for a multitude of service management aspects in distributed systems [262]. Furthermore, there are many stable implementations. Thus, a first step was the selection of an appropriate CORBA implementation, based on an evaluation of the runtime characteristics (see Subsect. 4.2.3). For an a-posteriori integration of non-CORBA-based tools (e.g. those which were designed for Microsoft Windows) with the CORBA infrastructure, Microsoft s COM/DCOM [847] interface is used and wrappers were developed to integrated those technologies (see Subsect. 4.2.4). 

At each internal tree node, a decision forest randomly can select F feature attributes, and evaluate just those attributes to choose the partitioning attribute. They tend to produce trees larger than trees where all attributes are considered for selection at each node, but different classes will be eventually assigned to different leaf nodes. At each internal tree node, RFs evaluate the quality of all possible partitioning attributes, but randomly select one of the Fbest attributes to label that node based on information gain, etc. RFs are an effective tool in prediction compared with boosting and adaptive bagging. 

---