"... In PAGE 15: ... IV. Substitution Patterns Table4 provides an illustration of the substitution patterns that are implied by each of the models. In particular, the table gives the probabilities from each of the four models under various scenarios compared with a base situation.... In PAGE 16: ...Table4 are not forecasts, the differences in the substitution patterns that arise under the different models will also occur in forecasting since these differences are intrinsic to the model specifications. We first describe the differences between the standard logit and the mixed logit in column 2 of Table 2, called mixed logit A.... In PAGE 16: ... We then describe differences with the pure probit and the mixed logit in column 4, called mixed logit B. In part 1 of Table4 , a mini electric car is introduced to a base situation consisting of five gas cars. The logit model, because of the iia property, implies that the new electric car will draw proportionately from all five of the gas cars.... In PAGE 17: ...relative to existing products, is potentially important in forecasting penetration rates for any new product, but especially for products that are expected to satisfy niche markets. For part 2 of Table4 , a second electric car in introduced, comparable in size to a gas subcompact. The previous scenario (five gas cars and a mini electric car) is taken as the base.... In PAGE 19: ... The log of the simulated probability is not unbiased for the log of the true probability; rather, given the log transformation, it is biased downward for a finite number of replications, with the bias decreasing as the number of replications increases. The figures in Table4 are consistent with these facts. Whether the bias can be considered large depends on the perspective that one takes.... In PAGE 20: ... We performed similar calculations for the pure probit. The results are given in the last column of Table4 . The variance in the average probability and the log-likelihood function is somewhat smaller for the GHK simulator with 50 replications than the mixed logit with the same number of replications.... In PAGE 21: ... Nevertheless, the recursive nature of the GHK simulator (where the range for the random draw for one alternative depends on the value of previous draws for other alternatives) is inherently slow compared to simulators, like the mixed logit simulator, which draw simultaneously from unrestricted ranges. In light of these issues, the results in Table4 are perhaps best interprested as simply an indication that the mixed logit simulator is reasonably accurate compared to the GHK simulator, particularly for given computer time.... ..."

"... In PAGE 10: ..., 2001). Table3 is based on the simulation results. In the column labeled analytical, we 260 give the throughput result from the GEM for each of the cases.... In PAGE 10: ... Based on the simulation output, we conclude that the 265 analytical results are reliable. 266 Table3 goes around here 5.... In PAGE 18: ...481 Figure 1: The three difierent blocking mechanisms 482 Figure 2: A zero-bufier queueing network representation 483 Figure 3: The generalized expansion method 484 Figure 4: Series topology 485 Figure 5: Split topology 486 Figure 6: Merge topology 487 Figure 7: GA optimization for earmarked symmetrical settings 488 Figure 8: GA optimization for earmarked asymmetrical settings 489 Figure 9: Queueing network structure 490 Figure 11: GA optimization for the fruit juice plant 491 Figure 10: Plot of throughput for difierent arrival rates 492 Figure 12: New queueing network structure 493 Table Legends 494 Table 1: Variables used in this paper 495 Table 2: Throughput for the symmetrical cases 496 Table3 : Simulation results for the symmetrical cases 497 Table 4: Throughput for the asymmetrical cases. 498 Table 5: Simulation results for the asymmetrical cases 499 Table 6: Processing rates used in the complex topology 500 Table 7: Data used in the complex topology 501 Table 8: Throughput rates for difierent conflgurations 502... ..."

"... In PAGE 3: ... Smelts were removed from the quot;other species quot; group and moved to the forage fish group beginning in 1999 as a result of fishery management plan (FMP) amendments 36 and 39 to the Bering Sea and Aleutian Islands and Gulf of Alaska groundfish FMPs. Individual other species known or suspected to occur in the Bering Sea and Aleutian Islands are listed in Table14 - 1. The species list was compiled from AFSC survey and fishery observer catch records, and is considered more comprehensive and up-to-date for the region than the general literature (Hart, 1973; Eschmeyer et al.... In PAGE 5: ... Maximum age reported for the longnose skate was 13 years, although there are many difficulties with ageing skates (Zeiner and Wolf, 1993). Little information is available on reproductive frequency in skate species, or on any Bathyraja species life history, but Table14 - 2 lists our best information on life history for all species in Alaska. Although little specific life history information exists for most skate species, they are generally thought to have limited reproductive capacity, and thus be vulnerable to overfishing (Sosebee, 1998).... In PAGE 7: ... FISHERY INFORMATION There is currently little directed fishing for species in this category in Alaska. Squid and other species are taken incidentally in target fisheries for groundfish, and aggregate catches of squid species (Table 14- 3) and the other species complex ( Table14 - 4) are tracked inseason by the Alaska Regional Office. Catch estimates by species group Because annual other species catches are reported in aggregate, catches by species group or individual species must be estimated using data reported by fishery observers.... In PAGE 7: ... The ratio of observed other species group catch to observed target species catch was multiplied by the blend-estimated target species catch within that area, gear, and target fishery. Total annual catch by species group has been relatively stable between 1997-2000 ( Table14 - 5). Estimated annual species group catches are reported by target, gear, and area in Tables 14- 6 through 14- 9 for 1997-2000.... In PAGE 8: ... Catch history for BSAI Squid and Other Species Squid are generally taken incidentally in target fisheries for pollock but have been the target of Japanese and Republic of Korea trawl fisheries in the past. Reported catches since 1977 are shown in Table14 - 3. After reaching 9,000 mt in 1978, total squid catches have steadily declined to only a few hundred tons in 1987-95.... In PAGE 8: ... In 1973-76 catches declined to a range of 33,000-70,000 mt annually as total catches of groundfish also declined. Catches of quot;other species quot; were relatively high from 1977-1981 (43,000-73,000 mt), but thereafter declined to a range of 5,000-13,000 mt in 1984-89 despite increased catches of total groundfish ( Table14 - 4). Part of the reason may be incomplete reporting of domestic catches before 1990.... In PAGE 9: ... It is clear that the AFSC bottom trawl surveys greatly underestimate squid abundance. Data from AFSC surveys provide the only abundance estimates for the various groups and species comprising the quot;other species quot; category ( Table14 - 10). Biomass estimates for the eastern Bering Sea are from a standard survey area of the continental shelf.... In PAGE 9: ... Stations as deep as 900 m were sampled in the 1980, 1983 and 1986 Aleutian Islands bottom trawl surveys, while surveys in 1991 and 1994 obtained samples only to a depth of 500 m. The actual catches made by research vessels are shown in Table14 - 11. Biomass estimates from AFSC surveys illustrate that sculpins were the major component of the other species complex until 1986, after which the biomass of skates exceeded that of sculpins.... In PAGE 9: ... However, sleeper sharks were the third highest CPUE on this pilot survey, indicating that they can be sampled by bottom trawls. This recent information suggests that it is the location and timing of the EBS trawl survey on the shelf during the summer, and not the use of bottom trawls for sampling which results in the apparently low biomass estimates for sharks in the EBS shelf ( Table14 - 10). Changes in distribution of particular species may also account for some of the biomass fluctuation of a group.... In PAGE 9: ... For instance, a cold water sculpin species, the butterfly sculpin (Hemilepidotus papilio), has been found to intrude into the northern portion of the survey area to a greater extent in some years than others, and accounts for some of the fluctuations in biomass of the sculpin group. Catch relative to biomass based on survey estimates Estimated skate and sculpin bycatch in the BSAI groundfish fisheries has ranged between 1-4% of their respective survey biomass ( Table14... In PAGE 10: ... If this is the case, current (1992-96) bycatch of grenadiers in the BSAI groundfish fisheries represents between 0.5 and 2% of the grenadier biomass in the BSAI region ( Table14 - 10). ANALYTIC APPROACH, MODEL EVALUATION, AND RESULTS Please see 1999 GOA SAFE Appendix E for a description of an experimental modeling approach and its results for GOA other species.... In PAGE 11: ... Several ABC and OFL options are available using the current tier 5 criteria for each species group within the other species category. Within tier 5, ABCs and OFLs are presented which are based on the most recent biomass estimate, the average biomass from the 1990 apos;s, and the average biomass over the entire survey time series for each species group ( Table14 - 13). These alternative ABCs and OFLs reflect our current understanding of the basic biology for each species group while protecting the less productive components of the category.... In PAGE 11: ... Bycatch of squid is reduced by limiting pelagic trawl fishing within relatively small areas of the shelf break; this has already been demonstrated through the indirect effects of closures related to Stellar sea lions. In 1999 and 2000, the pollock fishery was restricted or removed from one area of historically concentrated squid bycatch and squid catch was cut to less than half that observed in 1997-1998 ( Table14 - 1). Another option for bycatch reduction is the use of specialized gear.... In PAGE 18: ...TABLES Table14 - 1. Other species and squids in the Bering Sea-Aleutian Islands, by scientific and common name; compiled from the AFSC survey database RACEBASE.... In PAGE 19: ... Table14 - 1 Continued Scientific name Common name BSAI Other Category Malacocottus zonurus darkfin sculpin Hemilepidotus sp. Irish lord Hemilepidotus gilberti banded Irish lord Hemilepidotus spinosus brown Irish lord Hemilepidotus zapus longfin Irish lord Hemilepidotus hemilepidotus red Irish lord Hemilepidotus jordani yellow Irish lord Hemilepidotus papilio butterfly sculpin Archistes plumarius Triglops sp.... In PAGE 21: ... Table14 - 2. Life history information available for BSAI and GOA skate species.... In PAGE 22: ... Table14 - 3. Estimated total (retained and discarded) catches of squid (mt) in the eastern Bering Sea and Aleutian Islands by groundfish fisheries, 1977-2001.... In PAGE 23: ... Table14 - 4. Estimated total (retained and discarded) catches of other species (mt) in the eastern Bering Sea and Aleutian Islands by groundfish fisheries, 1977-2001.... In PAGE 24: ... Table14 - 5. Estimated total catch (t) of non-target species groups by FMP category and area, 1997-2000.... In PAGE 25: ... Table14 - 6. 1997 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 26: ... Table14 - 7. 1998 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 27: ... Table14 - 8. 1999 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 28: ... Table14 - 9. 2000 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 29: ... Table14 - 10. Estimated biomass (t) of BSAI squid and other species from various AFSC surveys.... In PAGE 30: ... Table14 - 11. Research catches of squid and other species in the BSAI, 1977-1998 (tons).... In PAGE 31: ... Table14 - 12. Published annual natural mortality (M) estimates for other species groups group species estimate reference squid Todarodes pacificus 0.... In PAGE 32: ... Table14 - 13. Potential BSAI ABC and OFL by species group for other species and grenadiers.... ..."

"... In PAGE 3: ... Smelts were removed from the quot;other species quot; group and moved to the forage fish group beginning in 1999 as a result of fishery management plan (FMP) amendments 36 and 39 to the Bering Sea and Aleutian Islands and Gulf of Alaska groundfish FMPs. Individual other species known or suspected to occur in the Bering Sea and Aleutian Islands are listed in Table14 - 1. The species list was compiled from AFSC survey and fishery observer catch records, and is considered more comprehensive and up-to-date for the region than the general literature (Hart, 1973; Eschmeyer et al.... In PAGE 5: ... Maximum age reported for the longnose skate was 13 years, although there are many difficulties with ageing skates (Zeiner and Wolf, 1993). Little information is available on reproductive frequency in skate species, or on any Bathyraja species life history, but Table14 - 2 lists our best information on life history for all species in Alaska. Although little specific life history information exists for most skate species, they are generally thought to have limited reproductive capacity, and thus be vulnerable to overfishing (Sosebee, 1998).... In PAGE 7: ... FISHERY INFORMATION There is currently little directed fishing for species in this category in Alaska. Squid and other species are taken incidentally in target fisheries for groundfish, and aggregate catches of squid species (Table 14- 3) and the other species complex ( Table14 - 4) are tracked inseason by the Alaska Regional Office. Catch estimates by species group Because annual other species catches are reported in aggregate, catches by species group or individual species must be estimated using data reported by fishery observers.... In PAGE 7: ... The ratio of observed other species group catch to observed target species catch was multiplied by the blend-estimated target species catch within that area, gear, and target fishery. Total annual catch by species group has been relatively stable between 1997-2000 ( Table14 - 5). Estimated annual species group catches are reported by target, gear, and area in Tables 14- 6 through 14- 9 for 1997-2000.... In PAGE 8: ... Catch history for BSAI Squid and Other Species Squid are generally taken incidentally in target fisheries for pollock but have been the target of Japanese and Republic of Korea trawl fisheries in the past. Reported catches since 1977 are shown in Table14 - 3. After reaching 9,000 mt in 1978, total squid catches have steadily declined to only a few hundred tons in 1987-95.... In PAGE 8: ... In 1973-76 catches declined to a range of 33,000-70,000 mt annually as total catches of groundfish also declined. Catches of quot;other species quot; were relatively high from 1977-1981 (43,000-73,000 mt), but thereafter declined to a range of 5,000-13,000 mt in 1984-89 despite increased catches of total groundfish ( Table14 - 4). Part of the reason may be incomplete reporting of domestic catches before 1990.... In PAGE 9: ... It is clear that the AFSC bottom trawl surveys greatly underestimate squid abundance. Data from AFSC surveys provide the only abundance estimates for the various groups and species comprising the quot;other species quot; category ( Table14 - 10). Biomass estimates for the eastern Bering Sea are from a standard survey area of the continental shelf.... In PAGE 9: ... Stations as deep as 900 m were sampled in the 1980, 1983 and 1986 Aleutian Islands bottom trawl surveys, while surveys in 1991 and 1994 obtained samples only to a depth of 500 m. The actual catches made by research vessels are shown in Table14 - 11. Biomass estimates from AFSC surveys illustrate that sculpins were the major component of the other species complex until 1986, after which the biomass of skates exceeded that of sculpins.... In PAGE 9: ... However, sleeper sharks were the third highest CPUE on this pilot survey, indicating that they can be sampled by bottom trawls. This recent information suggests that it is the location and timing of the EBS trawl survey on the shelf during the summer, and not the use of bottom trawls for sampling which results in the apparently low biomass estimates for sharks in the EBS shelf ( Table14 - 10). Changes in distribution of particular species may also account for some of the biomass fluctuation of a group.... In PAGE 9: ... For instance, a cold water sculpin species, the butterfly sculpin (Hemilepidotus papilio), has been found to intrude into the northern portion of the survey area to a greater extent in some years than others, and accounts for some of the fluctuations in biomass of the sculpin group. Catch relative to biomass based on survey estimates Estimated skate and sculpin bycatch in the BSAI groundfish fisheries has ranged between 1-4% of their respective survey biomass ( Table14... In PAGE 10: ... If this is the case, current (1992-96) bycatch of grenadiers in the BSAI groundfish fisheries represents between 0.5 and 2% of the grenadier biomass in the BSAI region ( Table14 - 10). ANALYTIC APPROACH, MODEL EVALUATION, AND RESULTS Please see 1999 GOA SAFE Appendix E for a description of an experimental modeling approach and its results for GOA other species.... In PAGE 11: ... Several ABC and OFL options are available using the current tier 5 criteria for each species group within the other species category. Within tier 5, ABCs and OFLs are presented which are based on the most recent biomass estimate, the average biomass from the 1990 apos;s, and the average biomass over the entire survey time series for each species group ( Table14 - 13). These alternative ABCs and OFLs reflect our current understanding of the basic biology for each species group while protecting the less productive components of the category.... In PAGE 11: ... Bycatch of squid is reduced by limiting pelagic trawl fishing within relatively small areas of the shelf break; this has already been demonstrated through the indirect effects of closures related to Stellar sea lions. In 1999 and 2000, the pollock fishery was restricted or removed from one area of historically concentrated squid bycatch and squid catch was cut to less than half that observed in 1997-1998 ( Table14 - 1). Another option for bycatch reduction is the use of specialized gear.... In PAGE 18: ...TABLES Table14 - 1. Other species and squids in the Bering Sea-Aleutian Islands, by scientific and common name; compiled from the AFSC survey database RACEBASE.... In PAGE 20: ... Table14 - 1 Continued Scientific name Common name BSAI Other Category Paricelinus hopliticus thornback sculpin Cephalopoda unident. cephalopod unident.... In PAGE 21: ... Table14 - 2. Life history information available for BSAI and GOA skate species.... In PAGE 22: ... Table14 - 3. Estimated total (retained and discarded) catches of squid (mt) in the eastern Bering Sea and Aleutian Islands by groundfish fisheries, 1977-2001.... In PAGE 23: ... Table14 - 4. Estimated total (retained and discarded) catches of other species (mt) in the eastern Bering Sea and Aleutian Islands by groundfish fisheries, 1977-2001.... In PAGE 24: ... Table14 - 5. Estimated total catch (t) of non-target species groups by FMP category and area, 1997-2000.... In PAGE 25: ... Table14 - 6. 1997 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 26: ... Table14 - 7. 1998 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 27: ... Table14 - 8. 1999 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 28: ... Table14 - 9. 2000 BSAI Squid and Other species and grenadier catches (t) by fishery and gear (1=bottom trawl, 2=pelagic trawl, 6=pot, 8=longline).... In PAGE 29: ... Table14 - 10. Estimated biomass (t) of BSAI squid and other species from various AFSC surveys.... In PAGE 30: ... Table14 - 11. Research catches of squid and other species in the BSAI, 1977-1998 (tons).... In PAGE 31: ... Table14 - 12. Published annual natural mortality (M) estimates for other species groups group species estimate reference squid Todarodes pacificus 0.... In PAGE 32: ... Table14 - 13. Potential BSAI ABC and OFL by species group for other species and grenadiers.... ..."

"... In PAGE 64: ... Table6 presents simulation and analytic results comparing performance of the vaxpy computation ordered under the assumptions of known and unknown alignment for a range of loop unrolling depths b. Assuming known alignment, access ordering improves perfor- mance over the natural reference sequence from 96% to 216%; under the assumption of unknown alignment performance is improved from 49% to 139%.... ..."

"... In PAGE 3: ... VOCALS is an international program that is part of the CLIVAR VAMOS (Variability of the American MOnsoonS) program. Participants in VOCALS currently include scientists from countries on the west coast of South America and the USA ( Table1 ). The overall goal of VOCALS is to develop and promote scientific activities leading to improved understanding, model simulations, and predictions of the southeastern Pacific (SEP) coupled ocean-atmosphere-land system, on diurnal to interannual timescales.... ..."

"... In PAGE 8: ... For the vast majority of spiralling cycles, this should be suf- FFcient because if the cycle was eventually to be lost, a diVTerent variable would eventually have to switch but would be very unlikely to wait longer than our integration time. The results of this survey are shown in Table2 . The numbers of occurrences of each behaviour type are out of 2000 network simulations.... ..."

"... In PAGE 8: ... This underscores the importance of having a good simulation model for DQDB, given the simplifying assumptions required by current analytical models. In Table1 we present our numerical results along with the results from the closed form analytical model obtained from Potter and Zukerman [6]. Inspection of Table 1 shows similar results when comparing the simulation runs versus the analytic results.... In PAGE 8: ... In Table 1 we present our numerical results along with the results from the closed form analytical model obtained from Potter and Zukerman [6]. Inspection of Table1 shows similar results when comparing the simulation runs versus the analytic results. The length of the simulation runs were set at 100 milliseconds (generating 20,000 slots).... ..."

"... In PAGE 7: ... This underscores the importance of having a good simulation model for DQDB, given the simplifying assumptions required by current analytical models. In Table1 we present our numerical results along with the results from the closed form analytical model obtained from Potter and Zukerman [6]. Inspection of Table 1 shows similar results when comparing the simulation runs versus the analytic results.... In PAGE 7: ... In Table 1 we present our numerical results along with the results from the closed form analytical model obtained from Potter and Zukerman [6]. Inspection of Table1 shows similar results when comparing the simulation runs versus the analytic results. The length of the simulation runs were set at 100 milliseconds (generating 20,000 slots).... ..."

"... In PAGE 1: ... However, the goals of providing a non-invasive, in situ, dense, long- term deployment of sensing infrastructure has necessitated that sen- sor nodes be cheap, wireless, battery-powered, and consume very little power. An unfortunate consequence of the limited resources of such nodes is that they are highly communication constrained, severely limiting deployment lifetime if raw data were transmitted to a central location ( Table1 ,[5]). For instance, as shown in Ta- ble 1, a micro-climate monitoring network composed of motes can be expected to last at most a couple of months even with optimized communication schedules.... In PAGE 3: ... Provid- ing a long-term storage and query processing capability requires storing summaries for long deployment periods. In storage-constrained networks ( Table1 ), however, resources have to be allocated for storing new summaries by discarding older ones. The goal of net- worked data management in our system is to discard summaries such that network storage resources are efficiently utilized, and grace-... In PAGE 6: ... In traditional wired sensor networks, the entire dataset would be available cen- trally, and could potentially be used to construct an optimal aging strategy using the above-mentioned constraint-optimization proce- dure1. Distributed scenarios such as wireless sensor networks have to op- erate with less information due to the overhead of centralized data collection ( Table1 ). In some scientific applications, a data gather- ing phase might precede full-fledged deployment (e.... In PAGE 9: ... This dataset provides a 15 x 12 grid of daily precipitation data for forty five years, where adjacent grid points are 50 kilometers apart. Both the spatial and temporal sampling are much lower than what we would expect in a typical sensor network deployment ( Table1 ). While the data sizes and scale would be expected to be larger in practice, this dataset has edges and exhibits spatio-temporal correlations, both of which are useful to understand and evaluate our algorithms.... In PAGE 9: ... Each node in the net- work would have 1095 samples to process every three years, enough to offer reasonable temporal compression benefit. In a typical deployment ( Table1 ), where nodes generate more data, the epoch would be much shorter. Hierarchy level (i) Num Cluster- heads (Nc) Compression Ratio Rate (ri) Total Data (Ncri) Raw 180 1 2190 ( ) 394.... ..."