### Table 5: Frequency of best results produced by individual degree-constrained heuristics on 2000 test networks.

1995

Cited by 38

### Table 4: Success rate of degree-constrained heuristics on 1000 test networks with a xed degree-constraint of 3.

1995

"... In PAGE 17: ...euristics. The maximum number of cases missed by any single heuristic was 7. To further verify the ability of the heuristics to nd solution with more stringent degree constraints, we also tested the heuristics on some of the graphs by imposing a xed degree- constraint of 3 for each node. Table4 summarizes the success rates of these simulations on 1000 test networks. As expected, both the number of cases in which backtracking was needed and the number of unsolved cases increased as a result of the small degree constraint.... ..."

Cited by 38

### Table 4: The size of the total discovered region of the AESSEA algorithms and mc-Prim on four mc-MST problem instances of increasing size. The column, jSPFPrj, gives the number of supported e cient solutions found by mc- Prim, using 1001 di erent vectors. The upper limit of the bounding box is determined, in each dimension, by taking the maximum value (over all algorithms) of any point on the combined Pareto front of each algorithm. The lower limit is taken to be zero in each dimension.

2001

"... In PAGE 7: ... However, mc-Prim apos;s is incapable of nding unsupported Pareto optima, of which there may be many between any two neighbour- ing supported optima. For these reasons, mc-Prim apos;s is able to display impressive results in Table4 , but will always fail to o er a ne-grained view of the tradeo surface, such as would be desired in many applications. Also, note that although mc-Prim apos;s is easily adaptable to degree-constrained problems, it certainly no longer guarantees to nd true optima in those cases.... ..."

Cited by 6

### Table 3: Summary of network flow based data aggregation algorithms

2006

"... In PAGE 19: ... This reduction in rates is achieved at the cost of increased network knowledge. Table3 presents the main characteristics and limitations of different network flow based data aggregation algorithms. Table 3: Summary of network flow based data aggregation algorithms ... ..."

Cited by 4

### Table 8. Comparison of temporal aggregation algorithms (n is the size of the base table)

2001

"... In PAGE 20: ... Compared to the basicSB-tree, the last three approaches require only a small, constant factor more storage and running time for their operations, and they are able to han- dle cumulative aggregates with arbitrary window offsets not known in advance. In Table8 , we compare our SB-tree algorithms and the endpoint sort algorithm of Sect. 5 with the other temporal aggregation algorithms discussed in Sect.... In PAGE 20: ... 2. For simplicity of presentation, Table8 provides only rough upper bounds on the running times of algorithms; please refer to the appropriate sections of this paper for detailed analyses. SB-trees, like most other algorithms compared in Table 8, do not handle the DISTINCT versions of SUM, COUNT, and AVG.... ..."

Cited by 56

### Table 5: The results of perimeter-degree constrained clustering formulation in terms of routed wirelength, place wire- length, the number of mildly congested edges and the number of severely congested edges are compared against the default placement and provided here as ratios. The last column provides the geometric mean of these ratios.

2004

"... In PAGE 20: ...2.1 is presented in Table5 . One of the significant observations is the reduction in both PlaceWL and RouteWL.... ..."

### Table 8 Comparison of temporal aggregation algorithms (D2 is the size of the base table).

"... In PAGE 18: ... able to handle cumulative aggregates with arbitrary window offsets not known in advance. In Table8 , we compare our SB-tree algorithms and the endpoint sort algorithm of Section 5 with the other temporal aggregation algorithms discussed in Section 2. For simplicity of presentation, Table 8 provides only rough upper bounds on the running times of algorithms; please refer to the appropri- ate sections of this paper for detailed analyses.... In PAGE 18: ... In Table 8, we compare our SB-tree algorithms and the endpoint sort algorithm of Section 5 with the other temporal aggregation algorithms discussed in Section 2. For simplicity of presentation, Table8 provides only rough upper bounds on the running times of algorithms; please refer to the appropri- ate sections of this paper for detailed analyses. SB-trees, like most other algorithms compared in Table 8, do not handle the BWC1CBCCC1C6BVCC versions of CBCDC5, BVC7CDC6CC, and BTCEBZ.... ..."

### Table 8 Comparison of temporal aggregation algorithms (D2 is the size of the base table).

"... In PAGE 18: ... able to handle cumulative aggregates with arbitrary window offsets not known in advance. In Table8 , we compare our SB-tree algorithms and the endpoint sort algorithm of Section 5 with the other temporal aggregation algorithms discussed in Section 2. For simplicity of presentation, Table 8 provides only rough upper bounds on the running times of algorithms; please refer to the appropri- ate sections of this paper for detailed analyses.... In PAGE 18: ... In Table 8, we compare our SB-tree algorithms and the endpoint sort algorithm of Section 5 with the other temporal aggregation algorithms discussed in Section 2. For simplicity of presentation, Table8 provides only rough upper bounds on the running times of algorithms; please refer to the appropri- ate sections of this paper for detailed analyses. SB-trees, like most other algorithms compared in Table 8, do not handle the BWC1CBCCC1C6BVCC versions of CBCDC5, BVC7CDC6CC,and BTCEBZ.... ..."

### Table 4-6. Comparison with Typical Asphalt Concrete Aggregate GradationVariability from Extraction Tests (adapted from Hughes 1996)

1998

"... In PAGE 30: ... Measurements were first taken from the prelevel surface to isolate the improvement of the overlay course. Table4 -1. Southbound Profilograph Summary Lane WheelPath Prelevel PRI [in/mi(mm/km)] Overlay PRI [in/mi(mm/km)] Improvementin PRI [in/mi(mm/km)] When Paved 1 Inside 4.... In PAGE 30: ...0(79) -0.8(13) D Table4 -2. Northbound Profilograph Summary Lane WheelPath Prelevel PR [in/mi(mm/km)] Overlay PRI [in/mi(mm/km)] Improvement inPRI [in/mi(mm/km)] When Paved 1 Inside 11.... In PAGE 35: ...erspective. When Markey et al. (1994) studied the initial impact of the WSDOT quality assurance (QA) specification, they included three QA projects. In Table4 -3, the statistical parameters for density from those jobs are compared to those from the mainline paving for the I-405 project. The I-405 project compares quite well.... In PAGE 35: ...aving for the I-405 project. The I-405 project compares quite well. Its mean average density is exceeded by only one of the other three projects (3522), and all but one of the other projects (3587) have greater variability. Table4 -3. Comparison of I-405 Average Rice Density Percentage Parameters to EarlyQA Projects from Markey et al.... In PAGE 36: ... Notwithstanding differences between measurement methods, the I-405 project variability was found to be distinctly less than those reported from the several sources included in Table 4-4. Table4 -4. Comparison with Reported Standard Deviation of Asphalt Concrete Air Voidsfor Roadway Compacted Mixtures (adapted from Hughes 1996) Source Year Method AV% I-405 1997 Nuclear 0.... In PAGE 36: ...roject was 6.8, which suggests a satisfactory period of service from the overlay. GRADATION Areas of quality performance that were explicitly specified in addition to density were the aggregates gradation and the asphalt cement content. Table4 -5 shows how favorably the mix of aggregate and asphalt cement fractions for this job compared to the job mix formula. For each aggregate size and for the asphalt cement content, job gradation fell well within the specified range limits.... In PAGE 37: ... Table4 -5. Gradation Summary for I-405 Project Item JMF% Job Average% Standard Deviation% One StandardDeviation% Specified Range Limits% 3/4 (19 mm) 100.... In PAGE 37: ...8 - 5.8 Table4 -6 helps to put the gradation variability into perspective. Aggregate gradation variability data published by Hughes (1996) is shown here for direct comparison to the I-405 gradation results.... In PAGE 38: ...The asphalt cement content variability can also be compared with some typical standard deviations. Table4 -7 shows how the variability from the I-405 project compares to data previously published from several sources and condensed by Hughes (1996). Only data from the last ten years are presented.... In PAGE 39: ... Table4 -7. Comparison with Typical Asphalt Cement Content Variability (adapted from Hughes 1996) Source Year Test Std Dev, % I-405 1997 Nuclear 0.... ..."

### Table 3. Current Consumption with/without data aggregation

"... In PAGE 7: ....5. Power Efficiency through Data Aggre- gation In addition to adaptive duty cycle management, pheromone-based data aggregation contributes to reduce power consumption of sensor nodes. Table3 compares the power consumption of Node 6 in the two configurations that agents perform data aggregation and do not. When agents do not perform data aggregation, agents use only energy level to decide whether they replicate themselves.... In PAGE 7: ... (They do not use pheromones.) Table3 shows that Node 6 consumes power 4.9% less when agents perform data aggregation.... ..."