【正文】 but also might influence the line items that were estimated based on the assumed state of the attribute. This, in turn, might cause a percentage escalation in the estimated project cost. An attribute is considered to be in the active state if, over the course of the project, the cost or status of an attribute differs from what was assigned to it at the estimating stage. For example, the labor productivity obtained during the course of the project might differ from what was assumed at the es timating stage. Similarly, at the estimating stage, a nonactive status might be assigned to the attribute, management, or project team. However, there is a possibility that during the course of the project the management or project team might make a decision error, influencing many other attributes. This would change the nonactive status of the attribute, management, or project team, to an active state. The probability and the resulting cost impact of these events cannot be neglected. Attribute state is defined by using a binary mode, where state = 1 implies that the attribute was in active state in that project, whereas state = 0 implies otherwise. The plex in terrelationship between the attributes suggests that even a minor change in the assumed equilibrium state of an attribute has the potential to trigger a domino effect. This effect could not only influence some other attributes but could also influence the project cost. Therefore, the binary mode of representation was considered to be most appropriate for this research, since any intennediate state between active and nonactive would not provide any additional infonnation. THE ATTRIBUTES For the purpose of this research, attributes that have a potential to cause project cost escalation were identified . In the past, several authors have examined the impact of isolated attributes on project cost However, no project management tool is available to account for the collective impact of all possible attributes. The attributes were divided into two groups, quantifiable and nonquantifiable attributes. Attributes that have a cost value associated with them in the project estimate were defined as quantifiable attributes, ., total material cost, total labor cost, total equipment cost, project management cost, and total cost of the project at end of work. Attributes that do not have a cost value associated with them in the project estimate were defined as nonquantifiable attributes. The need to differentiate between quantifiable and nonquantifiable attributes is elaborated later under modeling assumptions. FIG. 1. Example Influence Pattern Refers to the percentage cost escalation over the estimated project cost. To satisfy these requirements, a DSS such as COMPASS would be most suitable. MODELING ASSUMPTIONS The interrelationships between attributes, the resulting influence pattern, and the impact of attributes on the project cost have been structured by defining the five following modeling assumptions: Assumption 1 If an attribute, ., F (refer to Fig. 1) is influenced by a set of attributes, ., C and D, then the individual influence of the attributes in that set on F (., the influence of C on F and the influence of D on F) is considered to be independent, . p[(F n C)I(F n D)] =p(F n C) (Ia) :. p[(F n C) n (F n D)] 。 p(F n D) =p(F n C) (I b) ::::) p[(F n C) n (F n D)] = p(F n C) X p(F n D) (Ie) Assumption 2 All nonquantifiable attributes are conditionally dependent on their preceding attributes, ., a nonquantifiable attribute can attain the active state only if at least one of its preceding attributes is in the active state。 ., attribute F (refer to Fig. 1) can attain the active state (., F = 1) only if at least one of its preceding attributes C or D is in the active state (., C = 1 or D = 1). However, this constraint is not applicable for quantifiable attributes, ., X, Y, and Z (refer to Fig. I), because quantifiable attributes, apart from being influenced by their preceding attributes, are also directly related with certain line items (., quantifiable attribute total material cost would be related with material cost associated with various other line items), some of which might be influenced by other active attributes that would define the state of that quantifiable attribute (., total material cost) as active Assumption 3 Only the starting attributes, ., A and B (refer to Fig. 1), can be influenced by factors external to the system, whereas other attributes within the system can only be influenced by attributes preceding them in the influence pattern (refer to Fig. 1). The system represents all of the attributes included in the influence pattern Assumption 4 There is a probability that, although an attribute is in the active state, the attributes influenced by it might not get into the active state, ., C = 1 and D = 1 but F =0 (refer to Fig. 1) A corollary to assumption 4 would be that the active state probability of an attribute is a function of the independent influence of its preceding attributes, as defined in the influence pattern, ., p(F =1) =f{p[(C =1) n (F =1)], p[(D =1) n (F = I)]}. It is important to note that the accuracy of the active state probability of attributes is contingent upon the interrelationships defined in the influence pattern by the user. For example, if attribute F were influenced by a third attribute (say, H) in addition to C and D (as defined in Fig. 1), then p(F = 1) =f{p[(C =1) n (F = 1)], p[(D =1) n (F =1)], p[(H =1) n (F = I)]). However, s