"... In PAGE 15: ...ontaining faults. This minimal threshold is referred to as the critical threshold. After the critical level is decided, suspicious statements are thus highlighted by the heuristic. In Table4 , Rows c and b are the critical thresholds: the ratio of the rank of the critical level to the number of ranked levels and the ratio of suspicious statements within and below the critical level to statements involved in the heuristic (i.... In PAGE 15: ... Because H14 would consider results of other heuristics, a precise critical threshold is hard to define for it. In Table4 , Rows a and b for H14 indicate the number of predicate statements in a tested program divided by the number of the executable statements and the number of the statements highlighted by H1, respectively. Row c tells the effectiveness of using H14 to locate faulty predicate statements.... In PAGE 15: ... A unique threshold, which makes the suggested domain reasonably small and consistently contain faults, is highly desirable. In Table4 , critical thresholds for various heuristics in Row b range from 1% to 91%. A standard threshold cannot be decided within this wide scope.... ..."

"... In PAGE 15: ...ontaining faults. This minimal threshold is referred to as the critical threshold. After the critical level is decided, suspicious statements are thus highlighted by the heuristic. In Table4 , Rows c and b are the critical thresholds: the ratio of the rank of the critical level to the number of ranked levels and the ratio of suspicious statements within and below the critical level to statements involved in the heuristic (i.... In PAGE 15: ... Because H14 would consider results of other heuristics, a precise critical threshold is hard to define for it. In Table4 , Rows a and b for H14 indicate the number of predicate statements in a tested program divided by the number of the executable statements and the number of the statements highlighted by H1, respectively. Row c tells the effectiveness of using H14 to locate faulty predicate statements.... In PAGE 15: ... A unique threshold, which makes the suggested domain reasonably small and consistently contain faults, is highly desirable. In Table4 , critical thresholds for various heuristics in Row b range from 1% to 91%. A standard threshold cannot be decided within this wide scope.... ..."

"... In PAGE 15: ...ontaining faults. This minimal threshold is referred to as the critical threshold. After the critical level is decided, suspicious statements are thus highlighted by the heuristic. In Table4 , Rows c and b are the critical thresholds: the ratio of the rank of the critical level to the number of ranked levels and the ratio of suspicious statements within and below the critical level to statements involved in the heuristic (i.... In PAGE 15: ... Because H14 would consider results of other heuristics, a precise critical threshold is hard to define for it. In Table4 , Rows a and b for H14 indicate the number of predicate statements in a tested program divided by the number of the executable statements and the number of the statements highlighted by H1, respectively. Row c tells the effectiveness of using H14 to locate faulty predicate statements.... In PAGE 15: ... A unique threshold, which makes the suggested domain reasonably small and consistently contain faults, is highly desirable. In Table4 , critical thresholds for various heuristics in Row b range from 1% to 91%. A standard threshold cannot be decided within this wide scope.... ..."

"... In PAGE 21: ...0 COMPUTATIONAL EVALUATION OF THE TWO APPROACHES Our test problems are based on actual data representing the GBA; Table 2 itemizes primary characteristics. Response requirement scenarios {?t : t e T} are specified in Table3 ; each reflects a different spill defined by the type of oil, volume, and spill characteristics. Figure 3 depicts the GBA, including six storage locations and two areas that might be used for staging.... In PAGE 21: ... Figure 3 depicts the GBA, including six storage locations and two areas that might be used for staging. Table 2: Galveston Bay Area: Characteristics ______________________________________________________________________________ Number of Component Types : 30 or 31 Number of Component System Types : 9 Number of Response System Types : 42, 60, or 90 Number of Storage Locations : 5 or 6 Number of Critical Time Points : 5 or 10 Number of Staging Areas : 2 ______________________________________________________________________________ Table3 : Galveston Bay Area: Response Requirements: {?t : t e T} ____________________________________________________________ Critical Time Response Requirement Scenarios {?t : t e T} Point... In PAGE 26: ...1 2 SKIM_1 2 970 BOOM1 3 FREEPORT SKIM1 3 TEXAS CITY PUMP31 BAY TOWN LTANK 1 BAY TOWN BOAT12 BAY TOWN BOAT21 BAY TOWN 1 2 SKIM_1 2 970 BOOM1 3 LEFT BAY SKIM1 3 SAN LEON PUMP31 BAY ENTRANCE LTANK 1 BAY TOWN BOAT12 BAY ENTRANCE BOAT21 BAY ENTRANCE Total clean up capability = 8730 gallons/hour at critical time point 1 The capabilities (in Gallons/hour) prescribed to meet the clean up requirements of 8000, 15000, 25000, 35000, and 45000 (see scenario 3 of Table3 ) for the five critical time points are 8730, 15430, 27220, 37410, and 45090 (the degradation of capabilities was considered in computing the prescribed response capabilities and only the capability prescribed for time period 1 is tabled). These results allow us to characterize problem instances for which each solution method is best suited.... ..."

"... In PAGE 5: ... Table 6: Resource Utilization Rates for Critical Re- sources and Number of Patients (All scenarios) Scenarios Doc- tors Physio- thera- pists Speech thera- pists No. of ad- missions Base Case 98% 95% 107% 1034 Experiment 1 108% 108% 117% 1289 Experiment 2 97% 91% 101% 1034 Experiment 3 85% 88% 85% 1034 Experiment 4 86% 88% 87% 1443 The manpower requirements for the Base Case and experiments are to be found in Table7 below. Table 7: Resource Requirements for Each Experiment (Critical Resources Only) ... ..."

"... In PAGE 3: ....5m, use minimal amount of polymer, be insensitive to poor tuning, and require a smooth control action. All of the control strategies tested satisfactorily controlled the sludge blanket height and resulted in effluent suspended solids concentrations of less than 6mg/l. They all required considerably less polymer per month than the 2000l/month current dosage ( Table1 ). This difference can be attributed partially to density currents in the settler due to its design which erode the sludge blanket and require the operators to maintain the sludge blanket at 0.... In PAGE 4: ... For each control strategy, setpoints were increased by 10% and proportional factors were decreased by 10%. The cumulative exceeding of the critical sludge blanket height is given in the last column of Table1 . There is a clear difference in the sensitivity of strategies using the sludge blanket height as compared to those without it, as is to be expected.... ..."

"... In PAGE 3: ....5m, use minimal amount of polymer, be insensitive to poor tuning, and require a smooth control action. All of the control strategies tested satisfactorily controlled the sludge blanket height and resulted in effluent suspended solids concentrations of less than 6mg/l. They all required considerably less polymer per month than the 2000l/month current dosage ( Table1 ). This difference can be attributed partially to density currents in the settler due to its design which erode the sludge blanket and require the operators to maintain the sludge blanket at 0.... In PAGE 4: ... For each control strategy, setpoints were increased by 10% and proportional factors were decreased by 10%. The cumulative exceeding of the critical sludge blanket height is given in the last column of Table1 . There is a clear difference in the sensitivity of strategies using the sludge blanket height as compared to those without it, as is to be expected.... ..."

"... In PAGE 7: ... By using a larger value for the difference before deciding two runs are different, we can decrease the error rate for a given topic set size (because the discrimination power is reduced) [12]. Table6 gives the critical value required to obtain no more than a 5% error rate for a given topic set size. For the area measure, the critical value is the minimum difference in area scores needed.... ..."

"... In PAGE 6: ... By using a larger value for the difierence before deciding two runs are difierent, we can decrease the error rate for a given topic set size (because the discrimination power is reduced) [10]. Table5 gives the critical value required to to obtain no more than a 5% error rate for a given topic set size. For the area measure, the critical value is the minimum difierence in area scores needed.... ..."

"... In PAGE 7: ... By using a larger value for the difference before deciding two runs are different, we can decrease the error rate for a given topic set size (because the discrimination power is reduced) [7]. Table6 gives the critical value required to to obtain no more than a 5% error rate for a given topic set size. For the area measure, the critical value is the minimum difference in area scores needed.... ..."