Generator-based integrated-circuit design flows are crucial for meeting the aggressive system-on-chip development timelines demanded by rapidly changing modern workloads. Generators allow chip designers to develop complex solutions to classes of problems rather than individual instances, allowing significant design changes late in the development cycle and enabling incremental improvements to existing solutions. The philosophies of welcoming changing requirements and steady, incremental development have been embraced by the software development community for some time, but have only recently been incorporated into hardware development. While this adoption of proven software development philosophies has decreased the turnaround time of systems-on-chip, productivity has been limited by hard-to-automate tasks like physical design. Each generated design instance requires a new human-generated floorplan or other changes to the physical design flow, limiting the throughput of design space exploration by the available engineering resources. Automation of physical design is therefore critical for state-of-the-art generator-based system-on-chip design.

This work describes a series of generator-based integrated circuits manufactured in 28nm FD-SOI and 16nm FinFET, outlines the physical design challenges encountered in their development, and presents a physical design methodology purpose-built to solve these challenges. The integrated circuits presented include an 8192-point digital spectrometer in 28nm FD-SOI, a dual-core RISC-V vector processor with on-chip fine-grain power management in 28nm FD-SOI, a dual-lane RISC-V vector processor with a dedicated on-chip power management core in 28nm FD-SOI, an eight-core RISC-V vector machine in 16nm FinFET, and a 21-core RISC-V vector machine with a systolic array accelerator in 16nm FinFET. The eight-core chip achieves a state-of-the-art energy efficiency of 209.5 GFLOPS/W on a half-precision matrix multiplication (GEMM) kernel.

The physical design methodology presented uses a framework, Hammer, to provide reusable physical design deliverables by decoupling the design-specific, tool-specific, and technology-specific aspects of back-end design along with a novel floorplan generation framework for Chisel designs. This physical design methodology has been incorporated into the Chipyard framework, an open-source RISC-V system-on-chip development platform leveraging the Chisel hardware construction language. The floorplan generation framework allows Chisel programs, which generate RTL, to specify composable floorplans without modifying the original source code. The flow solves common challenges associated with floorplanning generated RTL, such as SRAM mapping and placement, demonstrating the efficacy of floorplan generation in reducing the overhead and cycle times of generator-based design.

To reduce transportation emissions and energy consumption, policymakers typically employ one of two approaches—changing technology or changing behavior. These strategies include demand management tools, such as ridesharing and vehicle control technologies—cleaner fuels and fueleconomy. Despite the benefits of a combined policy approach, these strategies are normally employed separately. Nevertheless, they have been linked occasionally, for instance in theelectric station car programs of the 1990s. Station cars are vehicles used by transit riders at the start or end of a trip.

In 1990, the California Air Resources Board (CARB) focused on reducing mobile airpollution by mandating that automakers introduce clean vehicles through its Zero Emission Vehicle (ZEV) Mandate. In 1998, significant flexibility was introduced through Partial ZEV credits for very low emission vehicles.

In 2000, CARB left the ZEV Mandate intact, but began considering new approaches,including station cars and carsharing. Carsharing is the short-term use of a shared-use vehicle fleet. In January 2001, recognizing the potential for station cars and carsharing to further improve air quality by reducing vehicle miles traveled—particularly with transit linkages—CARB proposed additional ZEV credits for vehicles in such programs. Thus, the mandate would formally link demand management and clean vehicles.This paper explores carsharing and station car developments, lessons learned, the ZEVmandate, and the proposed credit structure. Finally, the authors conclude with policy andresearch recommendations for enhancing the success and impact of this combined approach.

To reduce transportation emissions and energy consumption, policy makers typically employ one of two approaches—changing technology or changing behavior. These strategies include demand management tools, such as ridesharing and vehicle control technologies that involve cleaner fuels and fuel economy. Despite the benefits of a combined policy approach, these strategies are normally employed separately. Nevertheless, they have been linked occasionally, for instance in the electric station car programs of the 1990s. Station cars are vehicles used by transit riders at the start or end of a trip. In 1990, the California Air Resources Board (CARB) focused on reducing mobile air pollution by mandating that automakers introduce clean vehicles through its Zero-Emission Vehicle (ZEV) Mandate. In 1998, significant flexibility was introduced through partial ZEV credits for very-low-emission vehicles. In 2000, CARB left the ZEV mandate intact, but began considering new approaches, including station cars and carsharing. Carsharing is the short-term use of a shared-use vehicle fleet. In January 2001, recognizing the potential for station cars and carsharing to further improve air quality by reducing vehicle miles traveled—particularly with transit linkages—CARB proposed additional ZEV credits for vehicles in such programs. Thus, the mandate would formally link demand management and clean vehicles. Explored are carsharing and station car developments, lessons learned, the ZEV mandate, and the proposed credit structure. Finally, policy and research recommendations are discussed for enhancing the success and effect of this combined approach.

To reduce transportation emissions and energy consumption, policymakers typically employ one of two approaches – changing technology or changing behavior. These strategies include demand management tools, such as ridesharing and vehicle control technologies – cleaner fuels and fuel economy. Despite the benefits of a combined policy approach, these strategies are normally employed separately. Nevertheless, they have been linked occasionally, for instance in the electric station car programs of the 1990s. Station cars are vehicles used by transit riders at the start or end of a trip. In 1990, the California Air Resources Board (CARB) focused on reducing mobile air pollution by mandating that automakers introduce clean vehicles through its Zero Emission Vehicle (ZEV) Mandate. In 1998, significant flexibility was introduced through Partial ZEV credits for very low emission vehicles. In 2000, CARB left the ZEV Mandate intact, but began considering new approaches, including station cars and carsharing. Carsharing is the short-term use of a shared-use vehicle fleet. In January 2001, recognizing the potential for station cars and Carsharing to further improve air quality by reducing vehicle miles traveled – particularly with transit linkages – CARB proposed additional ZEV credits for vehicles in such programs. Thus, the mandate would formally link demand management and clean vehicles. This paper explores carsharing and station car developments, lessons learned, the ZEV mandate, and the proposed credit structure. Finally, the authors conclude with policy and research recommendations for enhancing the success and impact of this combined approach.

Most trips in U.S. metropolitan regions are driven alone, which is costly to individuals and society and leads to congestion and air pollution. A shared-use system aims to reduce traffic by reducing the number of cars needed by households and encouraging commuters to walk, bike, and use transit, at least for part of their trips. For commuters especially, shared-use vehicles could offer a low-cost, low-hassle alternative to private vehicles. Furthermore, carsharing should help air quality by incorporating low-emission vehicles into shared-use fleets.

In November 1997, the California Air Resources Board proposed modifying the Zero Emission Vehicle (ZEV) mandate such that certain vehicles with measurable tailpipe emissions would be allowed to earn partial credit toward the 10% requirement of 2003. This proposed change in the ZEV mandate would provide automakers with greater incentive to bring a broad range of very low-emitting vehicles to market, and would reduce the need to sell as many battery electric vehicles. Partial credits would be given to vehicles with very low tailpipe emissions, all-electric driving capability, and that use inherently clean fuels. Even very clean-burning gasoline vehicles could earn credits. This report describes the proposed methods and conditions for granting partial ZEV credits, along with illustrative examples. The implications of the proposed changes are analyzed, and the view of different stakeholders briefly characterized.

The CarLink field test combined short-term rental vehicles with communication and reservation technologies to facilitate shared-vehicle access. The ten-month demonstration was implemented and researched by two teams at the Institute of Transportation Studies at UC Davis. Project partners included American Honda Motor Company, BART, Caltrans, PATH, and LLNL. INVERS and Teletrac provided advanced carsharing and vehicle tracking technologies.

CarLink II was a commuter-based carsharing pilot project administered by the Institute ofTransportation Studies at the University of California, Davis (ITS-Davis) in conjunction with Caltrans, American Honda Motor Company, and Caltrain. Partners for Advanced Transit andHighways (PATH) researchers conducted the evaluation. Pilot objectives included testing an advanced carsharing system, understanding user response to this service, and testing its long-term sustainability. From July 1, 2001 to June 30, 2002, the CarLink II program was deployed in the field and continued the investigation of commuter-based carsharing originally explored in the1998 CarLink longitudinal survey and the 1999 CarLink I field test. Lessons learned during the CarLink I field test helped guide the project team's design of the CarLink II project, resulting inseveral differences and improvements.

Chapter One: U.S. Shared-Use Vehicle Findings: Opportunities and Obstacles for Carsharing & Station Car Growth

Shared-use vehicle services provide members access to a fleet of vehicles for use throughout the day, without the hassles and costs of individual auto ownership. From June 2001 to July 2002, the authors surveyed 17 U.S. shared-use vehicle service organizations on a range of topics, including organizational size, strategic partnerships, pricing strategies, insurance costs, and technology applications. While survey findings demonstrate a decline in the number of organizational starts in the last year, the rate of operational launches into new cities, membership, and fleet size continue to increase. Several growth-oriented organizations in the U.S. are responsible for the majority of this expansion. The authors explore several factors that challenge shared-use vehicle growth, such as high capital investment (or start-up costs), dramatic hikes in insurance rates, and scarcity of cost-effective technologies.

The authors conclude that while early niche market findings are encouraging, the ability of this emerging sector to actualize its total environmental, economic, and social goals may be limited without the collective support of private industry (e.g., automakers, insurance providers, technology producers), public agents (e.g., transit and governmental agencies), and shared-use vehicle programs. Indeed, public-private partnerships and cooperation among shared-use vehicle providers may play a key role in addressing insurance and technology costs and assuring the long-term viability of this market.

Chapter Two: A Framework for Testing Innovative Transportation Solutions: A Case Study of Carlink—A Commuter Carsharing Program

Transit accounts for just two percent of total travel in the U.S. One reason for low ridership is limited access; many individuals either live or work too far from a transit station. In developing transit connectivity solutions, researchers often employ a range of study instruments, such as stated-preference surveys, focus groups, and pilot programs. To better understand response to one innovative transit solution, the authors employed a number of research tools, including: a longitudinal survey, field test, and pilot program. The innovation examined was a commuter carsharing model, called CarLink, which linked short-term rental vehicles to transit and employment centers. Over several years, researchers explored user response to the CarLink concept, a field operational test (CarLink I), a pilot program (CarLink II), and a commercial operation (the pilot was turned over to Flexcar in summer 2002). This multi-staged approach provided an opportunity for researchers to learn and adapt as each phase progressed. In this paper, the authors outline the CarLink model, technology, and early lessons learned; describe CarLink II operational understanding; provide a synopsis of the pilot program transition; and offer recommendations for future model development.

Chapter Three: Travel Effects of a Suburban Commuter-Carsharing Service: A Carlink Case Study

Since 1998, carsharing programs (or short-term auto rentals) in the U.S. have experienced exponential membership growth. As of July 2003, 15 carsharing organizations collectively claimed 25,727 members and 784 vehicles. Given this growing demand, decision makers and transit operators are increasingly interested in understanding the potential for carsharing services to increase transit use, reduce auto ownership, and lower vehicle miles traveled. However, to date, there is only limited evidence of potential program effects in the U.S. and Europe. This paper presents the travel effects of CarLink?a commuter carsharing model with explicit links to transit and employment in a suburban environment?in the context of participant demographic and attitudinal market profiles. A variety of research methods (including focus groups, interviews, questionnaires, and travel diaries) captured the following commute travel effects from the CarLink I and II programs:

* Increased commuter rail mode share by 23 percentage points in CarLink I and II; * Reduced drive-alone mode share by 44 and 23 percentage points in CarLink I and II, respectively; * Decreased average daily vehicle miles traveled by 23 miles in CarLink II and by 18 miles in CarLink I; * Increased travel time but reduced stress; * Reduced vehicle ownership by almost six percent in CarLink II; and * Reduced parking demand at participating train stations and among member businesses.

The typical CarLink I and II member was more likely to be highly educated, in an upper income bracket, and professionally employed than average Bay Area residents. CarLink I and II members also displayed sensitivity to congestion, willingness to experiment, and environmental concern. The travel results of CarLink I and II are compared to those of neighborhood carsharing models in the U.S. and Europe to suggest the importance of CarLink's explicit transit and employment connections and the value of carsharing in a suburban location.

Chapter Four: Applying Integrated ITS Technologies to Carsharing System Management: A Carlink Case Study

Carsharing is the short-term use of a shared vehicle fleet by authorized members. Since 1998, U.S. carsharing services have experienced exponential growth. At present, there are 13 carsharing organizations. Over the past three years, electronic and wireless technologies have been developed that can facilitate carsharing system management in the U.S., improve customer services, and reduce program costs. This paper examines the U.S. carsharing market; the role of advanced technology in program management, including CarLink lessons learned; and technology benefits to this nascent market.

Chapter Five: Carlink—A commuter Carsharing Model: Conditions for Economic Viability

At present, only a few alternatives exist to facilitate transit access. To expand the suite of viable modes, more demand-responsive mobility services should be developed. Under specific conditions, short-term rental vehicles linked to transit (or commuter carsharing) can offer a sustainable transportation alternative to private vehicles, particularly in suburban locations where transit connectivity is more limited. From January to November 1999, a commuter carsharing model, called CarLink I, was tested in the East San Francisco Bay Area. Building on CarLink I, the CarLink II pilot program was deployed from July 2001 to June 2002 in the South San Francisco Bay Area with the option of transitioning to a third-party provider. Examination of CarLink I and II economic data and scenario analyses—which explore program modification effects on viability—revealed several economic success factors. In addition to fixed monthly CarLink rates, the introduction of hourly rentals is an important strategy to diversify CarLink income streams and maximize vehicle use, as well as a revenue-risk sharing approach among partner participants (e.g., employers, transit providers). Additional strategies to create more viable commuter carsharing services include raising rates for business customers, lowering insurance rates, controlling costs through technology and scale, and marketing strategies.