AMBUILDERS 5a7c315affe00d0bd44e7b70 False 8 7
Space Utilization Op


Services INR  0 INR  0
  • S
  • M
  • L
  • 1
  • 2
  • 3

Space Utilization Optimization Overview Effective space management relies on information about people, places, and processes. Geographic information system (GIS) technology helps facility managers organize and spatially visualize where and in what type of space people work. The GIS team at the National Aeronautics and Space Administration (NASA) Langley Research Center (LaRC) in the city of Hampton, Virginia, has used GIS to handle the mammoth task of its facility's space management across its 800-acre facility, which holds approximately 400 buildings and test structures totaling 3.7 million square feet. Its 6,700 rooms house 4,000 employees. The center is primarily identified with wind tunnel research but supports many other disciplines, including structures and materials, flight electronics, and atmospheric science. Because of LaRC's diverse activities, the infrastructure is massive and complex, with a variety of facilities. This information is based on the LaRC GIS team's experience at NASA as well as experience gained working with partners that have similar goals. The space utilization optimization and supporting tools were developed to address the need for optimizing facility usage to minimize operational costs while maximizing synergy between employees and organizations. As these tools were developed for relatively complex government facilities, they should be readily adaptable for use in other environments. The GIS team's long-term goal is to enhance and develop this technology in order to provide objective tools for organizations to control cost while accomplishing their primary mission with increased efficiency and effectiveness. Background LaRC, the oldest of 10 major NASA centers, is located in Hampton, Virginia, adjacent to Langley Air Force Base. Historically, the general scope of LaRC includes 800 acres and 300 plus buildings comprising 6,000 plus rooms totaling more than 3 million square feet, with assets valued at approximately $3 billion. The facilities were designed to house more than 4,000 civil service and contract employees. LaRC has been identified with aeronautical or wind tunnel research for over 50 years but also supports many other disciplines including structures and materials, flight electronics, and atmospheric sciences. Such diverse capabilities require a massive and complex infrastructure as well as specially designed buildings. More recently, LaRC has experienced significant reductions in operational, maintenance, and staffing funding while simultaneously adapting to major missio Space Utilization Optimization J-9780 June 2009 2 LaRC first floor building interior detail superimposed on an aerial photograph. Bringing in just the first floor plans for the buildings is too dense at full extents of the map for staff to use the information effectively. GIS for LaRC was born more than two decades ago out of an anticipated need to address facility planning more efficiently and effectively by capturing corporate knowledge into configuration management and decision support systems. Integration of datasets was seen as a key method to ensure sustainable data maintenance. Integration efforts include support to master plan, real property, utility system, environmental, facility full cost, maintenance, personnel, and space utilization functions. A primary goal was to cultivate and manage the most stringent data that was available, reducing the number of datasets and encouraging data use in order to simplify maintenance by multiple users. GIS has been fostered at LaRC by a team consisting of a few civil servants with contract support making up the majority of personnel. The team has grown in recent years to approximately a dozen, plus a year-round group of about 10 interns. The GIS team has strived to be associated with LaRC's Facility Engineering group so that the philosophy of "most stringent data" was supported. The GIS team not only addresses the needs of LaRC but also outside organizations through partnerships. These partnerships were formed when outside organizations, such as other NASA centers or government entities such as the Air Force, were interested in a capability that LaRC's GIS team was pursuing. These partnerships allowed distributed funding for development efforts, which allowed more aggressive pursuit of many capabilities over the years. Building Interior Data LaRC's GIS team manages plant-level spatial data for the facility, as well as having managed building interior details in GIS for more than a decade. The LaRC GIS team has chosen to maintain room-level data in a nongeoreferenced fashion. The data looks much like an architect's plan for a building that depicts multiple floors of a building orthographically laid out on a large drawing. The large drawings for all the buildings are then maintained in a grid that is independent of details for other buildings and infrastructure such as roads. Subsequently, the data is translated, scaled, and rotated to allow the room-level data to overlay the plant-level data when needed. To accomplish the translation of data, diagonals are used to match the data maintained in the grid to the georeferenced data. Space Utilization Optimization J-9780 Esri White Paper 3 A valuable application of this technique allows users to demonstrate outdoor/indoor utility alignment. This approach maintains legacy views of building interior space as well as deals with known distortions between building and plant data. Additionally, the process has been extend Space Utilization Optimization J-9780 June 2009 4 Example of Need for Space Utilization Optimization LaRC, along with the rest of the agency and other federal and state organizations, as with much of the corporate world, continues to experience pressure to undergo major downsizing and reorganization. The stimulus for this includes alterations in mission requirements, the introduction of full-cost accounting methods, funding cuts, and the excessive operational and maintenance costs associated with aging infrastructure. LaRC felt even further pressure from the change in NASA's mission emphasis from aeronautics to space exploration. As a reaction to external stimulus, LaRC established an initial goal to continue to support projected missions with reductions of 25 percent in land, buildings, and personnel at LaRC. The extreme complexity of LaRC's facilities (partially because they are over 50 years old—heavy walls from Cold War era construction) was a major motivation for LaRC to pursue automated optimization for space utilization. Additionally, there is a need for a mechanism to overturn the heavily entrenched legacy processes of "perceived ownership," whereby an organization or function that had existed in a building for decades essentially determined how the building would be used forever. As with other GIS processes, LaRC strives to provide objective decision support tools to dislodge the legacy ownership environment and to have reorganization efforts driven more from potential synergy benefits and cost avoidance. The current approach reports the value in dollars of any scenario under consideration in an effort to reduce the impact of subjective decision making. The need existed for a centerwide strategic capability to support more effective and efficient facility management through the use of GIS and optimization technologies. In addition to the usual requirements for office space, NASA centers and other federal facilities (e.g., laboratories, wind tunnels, and launch facilities) must satisfy a wide range of special space requirements to support a diverse and evolving mission. Efficiently managing the changing mission and projects with limited resources is a challenging task. Space and facilities managers needed a decision support system that could track changing workforce and project requirements and map those against available facility resources. Such a system will significantly enhance decision making in determining optimal space allocation and scheduling by considering project lifetimes and changes in resource needs while balancing immediate needs with forecasted requirements. Additionally, the system would provide objective analysis to achieve an optimal combination of adaptive reuse and "repair by replacement" projects to address deteriorating conditions and increasing maintenance backlog for LaRC's facilities. In late 2004, LaRC was preparing for a major reorganization and bracing for four or more months of turmoil. During this period, it expected to relocate up to 3,000 people, reduce average office space per person from more than 190 square feet currently to a target 125 square feet and free up approximately 100 facilities for closure and demolition. This situation was the mother of invention for automated space optimization at LaRC. With the challenge looming, LaRC's GIS team began the development of a series of automated tools based on GIS and RDBMS technology to support managers in this complex reorganization effort. These tools support the development of multiple scenarios using both objective and subjective decision criteria to visualize and analyze various possible space allocation solutions. LaRC's GIS team has learned much from the initial endeavor and is pursuing a project to refine the previously developed prototype capabilities, which will result in an integrated and sustainable space utilization decision support environment for LaRC and other large and complex facilities. Space Utilization Optimization J-9780 Esri White Paper 5 Initial Capabilities The capabilities outlined in this section represent deliverables under the earliest phase of the evolutionary development strategy. During this early phase, visualization of current personnel location, organizational distribution, and space utilization was made readily available to LaRC personnel with standard Web browsers. This tool allowed infrastructure managers to readily assess current organizational space allocation and to determine overcrowded and/or underutilized facilities. To help meet the challenge of reducing operational costs by more efficiently utilizing available space, the LaRC GIS team tested and continued the development of optimization algorithms. Originally developed in conjunction with NASA design optimization engineers, the initial algorithm was designed to help redistribute organizational slots based on a variety of user-defined criteria (e.g., lab/technical space constraints, organizational synergy constraints, move minimizations). A Web-based tool was originally developed and is expected to be reengineered to assist space utilization planners in analysis of information concerning laboratory and technical space. Data collected from this tool was made available to the optimization algorithm to further refine constraint definition and cost metrics. Additional improvements in efficient space utilization were explored through the development and implementation of more advanced optimization algorithms, such as genetic or simulated annealing algorithms, for space optimization as a part of the proposed space planning framework. This image displays where the center's personnel are located near their labs. It is easy to see in this image that there is a fragmentation of organizations and an opportunity for better use of space. Using the power of ArcGIS® software coupled with custom Visual Basic® code, the LaRC GIS team developed an early GIS application that allowed space utilization managers to construct and evaluate various what-if move planning scenarios. The application allowed the interactive manipulation of organizational slots both within buildings and between buildings while displaying space utilization parameters (e.g., over/under capacity) in real time. Coupled with the optimization algorithm, this tool Space Utilization Optimization J-9780 June 2009 6 enabled space utilization managers to rapidly evaluate proposed scenarios. The tool retains the legacy capability for managers to add constraints such as room or personnel "lock down," after which the remaining space can be optimized. Space Utilization Optimization Process and Tools Space allocation planning is a complex problem involving the allocation of limited resources to meet business goals, reduce operating costs, and promote an effective and productive workplace. The optimization process has many facets and is very complex. So that some benefit could be realized from incremental accomplishments during the development phase, the effort was broken down into modules. These modules include visualization, optimization algorithm, data maintenance, Web interface, and technical space. The following narrative roughly parallels the development process. Visualization One of the difficulties with talking about space planning in the context of a large organization like NASA's LaRC is the problem of getting the big picture without losing the necessary details. The following outlines the next generation of the user interface. This interface will employ a dashboard concept that will allow the user to access any or all data representations such as the unit square diagram, the plant-level map, the building interior layout, and any supporting tabular data. The capability will allow consumers of solutions to readily see relative size and proximity of buildings, rooms, and personnel for an entire base. Additionally, the tool uses any symbolization and labeling available through GIS. For more detailed analysis, additional conventional map views and building layouts with room details are available. The user will be able to visualize current conditions and various proposed optimization solutions and manually adjust conditions through tools such as drag and drop. Thus far, GIS-based tools that address both plant- and individual building-level visualization have been developed. To address plant-level visualization, a tool referred to as the unit square or spatial subdivision diagram was created that uses an abstract representation of all buildings at LaRC compressed into a rectangle. Since the focus is on space utilization, features such as roads, parking lots, and grass are not addressed in the data-viewing tool. The diagram capability maximizes any view window for the data under consideration. The tool effectively eliminates white space (distance between buildings) on a drawing by compression and proportioning the data. Bold rectangular areas represent all of the facility's buildings. They are roughly oriented based on proximity to buildings around them. The size of the polygon represents usable space within the facility. The rectangle that represents a building can be further subdivided to address all rooms. The blocks within the building represent the space of each room. Again, size is proportional and relative position to the real-world location is maintained, with multiple floors of buildings delineated with shadow lines. There are many areas in a building, such as circulation areas, stairwells, bathrooms, and mechanical equipment rooms, that may not be germane to a space allocation problem and thus need not be included for analysis via the visualization tool. Symbolization can then be applied to the unit squares for buildings and rooms to show those spaces that are appropriate for general office use and technical areas symbolized based on the owning organization. The closeness in color is indicative of the closeness of the organizations in the organizational hierarchy. A similar scheme is used to color point features indicating all the personnel at LaRC. The aforementioned technique can visualize not only area but also any commodity or value, such as maintenance cost, where size could represent cost. The goal for optimum synergy would be to have similar colors in close proximity. The unit square diagram provides a dense and concise visualization, allowing user interaction Space Utilization Optimization J-9780 Esri White Paper 7 with the massive and diverse data associated with managing a complex facility's space. The unit square approach can be extended to illustrate each person's space at LaRC. Thus, from one interface or data view, the user can interact with data for the entire center that represents either building-, room-, or personnel-level data. Metrics and Constraints Due to the complexity of the space utilization planning task, it is useful to have a model so that one scenario can be compared to others as progress is estimated toward some goal. That model must address critical variables and effects but be simple enough to make analysis feasible. Infrastructure-related costs are obvious components to be addressed in the model; lessobvious components of a cost model are those that capture the more nebulous effects on performance of a group such as organization communications. These communications or organizational interactions translate to a synergy component in the overall equation. For the model to be as flexible as possible, it needs to be made as general as possible. The first generalization is to refer to any person or function that consumes space as a consumer. This could be a laboratory, a conference area, and so forth. Similarly, both office and technical areas are simply referred to as space. These spaces provide certain resources that the consumers need. The most common resource is, of course, area; however, additional resources can be modeled like communications jacks, bandwidth, power, and environmental impact. Once the basic components are determined, the ability to evaluate and compare particular allocation plans is needed. Questions that need to be asked include Does this allocation meet our rules? and In comparing two valid allocations, which one is objectively better? Processing the optimum placement of consumers in LaRC's facilities will be governed by constraints. These go/no go parameters include considerations such as adequate space for the function (minimum area for various types of employee), compatibility with adjacent functions (supervisors/employees not placed in the same room), and compatibility with features that the space readily provides (floor loads, high bay, etc.). Further evaluation of the quality of the proposed use of space is driven by metrics. These softer controls address issues such as organizational synergy (the closer personnel are positioned within an organization, the better the metric), moving costs, and symmetrical distribution of space. The result is a proposed scenario that meets all constraints and is rated by metrics. The quality of the solution is expressed in dollars, allowing space utilization personnel to present various solutions to management for consideration. The different scenarios will inevitably include manual adjustments to address an organization's political issues; the cost of the potential changes provides a discriminator for more objective analysis. Management may then decide if the manually induced change is worth the expenditure to the organization to address the political issue. Data Management As with any system, collecting and maintaining accurate and current data represent challenges for the organization. One problem is that the aforementioned process requires pulling data from many different sources, including personnel, GIS, and space utilization, with these sources of data constantly changing. The process used is to snapshot the data; resolve any consistency issues; and, as needed, reconcile any resulting plan with a current snapshot. Resolving and mapping the source data to the general model are very time intensive and involve both data and scenario-specific considerations. An XML schema was developed to provide a language for the model. This language is tightly bound to the source code implementation and is reflective of the very general Space Utilization Optimization J-9780 June 2009 8 nature of the model. Since virtually every data source has some sort of mapping to an XML schema, the XQuery language was used to leverage data transformation. The result is a "recipe" that addresses the "heavy lifting" required to arrange or systematize the corporate knowledge. Before the XML/XQuery capability was available, preparation for an optimization run took 2 weeks of intensive effort. Using the recipe capability, preparation for a run takes about 30 minutes. Optimization Algorithm With the model instance, allocations can be manually manipulated to get feedback on the constraints and metrics; however, having the computer automatically find an allocation that is in some sense optimal is preferable. To get a grasp of how big a possible solution set for a complex facility is, it's better to start with a small model such as a proposed move involving four people and five rooms. In this example, there are many possible mappings of the consumers to the space. In fact, for this simple problem, there are about a thousand distinct allocations. Each could be easily calculated, the ones that don't meet the constraints can be thrown out, and the remaining one that has the lowest composite cost metric can be selected. When the challenge LaRC faces (with roughly 4,000 people and 4,000 spaces) is addressed, the resulting number of permutations is enormous. This equates to 1014,400. To get a handle on how large this number is, an estimate of the number of grains of sand in all the world's beaches is 1018 and the number of atoms in the observable universe is 1080. Obviously, no amount of computational power could ever force an answer via an exhaustive process. What is needed is a method that gives a very good result in a reasonable amount of computational time. The current optimization approach divides the search into two parts. The first part is a constraint solver that takes an allocation, which has violated one or more constraints, and finds a solution in the same neighborhood that satisfies all the constraints. The design is interesting because what was needed was a framework that is extensible, so the process needs to work for constraints not yet imagined. The second part of the search is a greedy heuristic that takes the most direct route from a given constrained solution to a constrained local optimum. It does this by iterating over a priority queue of consumers and using an efficient local first search. The solution obtained is a local, not a global, optimum. Optimization is conducted not only from the current condition but also from a random allocation. The results from these processes typically are good synergy within the smallest organizational units, but a lot of issues are noted where large-scale changes might improve the higher-level synergy and collocation with technical space. What is preferred is to take the good ideas from each and make them fit together. By evaluating the metrics per consumer in each solution, a new solution can be created that tries to combine the best of both. Doing this inevitably results in a solution that doesn't meet the constraints and may not initially look better, but through constraint solving and reapplying the greedy heuristic, a facility has a viable alternative. Combine this with a progressive filtering algorithm, and improvements can often be found to any local optimum. Feeding the process a stream of random solutions enables it to integrate progressively smaller ideas into an improved final solution. The following illustrates the benefits of large-scale optimization at a complex facility such as LaRC. The combined metrics for LaRC's current space allocation produces a value around $25 million. If the greedy heuristic alone is applied, requiring just a couple of minutes, the resulting allocation has a combined metric of $15 million; if it's run Space Utilization Optimization J-9780 Esri White Paper 9 through the metaheuristic over about 24 hours, the resulting solution is approximately $11 million, and the rate of improvement is generally very low after that. Much of the previously described effort has been documented through videos that are available from LaRC's GIS Web site at Future Efforts Web Interface Recent work is focusing on the development of a dashboard-style Web interface that affords an information-dense visual analysis experience. The capability gives contextsensitive and immediate graphic feedback, showing progress in meeting constraints and reducing costs, thereby offering space planners a powerful decision support system. The current focus is on providing drag-and-drop-style planning for fine tuning of optimization solutions. Administrative Space In one example, the optimization separated staff functions from their respective senior managers, based on the lack of definition of synergy between the functions. Subsequent analysis of this problem may lead to a new way to describe affinity between managers and their immediate support staff and the type of space they need. Under this new model, office suites (areas compatible for senior manager, deputy, and administrative assistant) and potential occupants may be assigned exclusivity values that will allow the optimization algorithm to better manage the unique resources (both real property and human). Optimization Algorithm Alternative search methods include population-based and hierarchical metaheuristics. Specifically, LaRC is working with Operations Research faculty at the College of William and Mary on extending the greedy heuristic with tabular search. Technical Space While the code fully supports technical space, there are a lot of very specific requirements that need to be mapped. Parallel development efforts are under way to extend automated optimization to address this diverse space. Currently, this process is focused on LaRC data, with an attempt to align with General Services Administration (GSA) categories, then to describe unique features associated with each category of technical space. Subsequently, average costs for demolition and creation of unique features will be used to calculate feasibility of converting one type of technical space to another. Scheduling The goal is to provide a scheduling capability that optimizes support to mission and projects and replaces the legacy model where a particular organization owns specific buildings. The Optimization tool will provide space utilization and facility managers with an immediate as well as a long-term overview of localized and broader options and impacts associated with current and proposed use of the facility's space. Staging and decisions regarding which project should be supported at the facility are planned to be addressed. The goal is to objectively support agency-level decisions regarding placement of future projects. The Optimization tool will facilitate analysis to support facility closures and leases on and off LaRC's site. This concept may be extrapolated to other organizations regarding functions such as mission changes for the facility or Base Realignment and Closure (BRAC). Automatic Routing The current distance model is Euclidean between the room and building centroids with a fixed interfloor distance. The GIS team would like to be able to model using more accurate distances derived from automatic routing within the facility to improve the solutions. Space Utilization Optimization J-9780 June 2009 10 Human Factors A very interesting area is social networks where many similar types of networks appear to follow a power law or scale-free structure. In these networks, most nodes have few connections, but some nodes have many. In LaRC's case, this would imply that some people within the organization are critical pathways for communication. Spin-offs In the process of implementing the space allocation planning tools, extensive knowledge and experience were gained with many new software technologies including XQuery, REST and AJAX Web architectures, the Dojo Toolkit, and Esri's ArcGIS for Server and JavaScript™/REST APIs. This will yield many beneficial products such as more intuitive and powerful Web interfaces for many GIS and RDBMS tools. Return on Investment In terms of tactical investment strategy, the proposed system promotes the public good at any facility by modeling and objectively analyzing impacts on issues such as safety and security, historic preservation, natural resources conservation, workforce productivity, environmental quality, and energy and materials efficiency. Enabling sound stewardship of public assets is a fundamental goal of such a strategic planning effort. The space utilization optimization system, its underlying technology, or spin-offs from development of the technology may benefit other federal facilities such as the Department of Defense (DoD), GSA, and the GIS industry in general through technology transfer activities. Use of optimization algorithms in concert with GIS promises extensive benefits for government facilities. The benefits associated with the system under development can be illustrated with a concrete example: The goal of a 2005 office consolidation project at LaRC was to reduce office space utilization from 190 to 125 square feet/person. A manual decision process was used due to time constraints and lack of funding for GIS support. The resulting subjective solution moved 2,500 personnel and achieved 149 square feet/person with an average cost per move of $354/person. Subsequently, a prototype automated optimization model similar to the one proposed here gave an objective solution whereby the goal of 125 square feet/person could have been achieved by relocating only 1,200 personnel. The optimization approach would have saved LaRC about $460,000, minimized disruption of normal operations, and achieved greater organizational synergy. This example addresses only move-related costs for a single effort focused on office space. Annual cost savings associated with additional facility closures accrued over multiple years and across multiple centers would result in millions of dollars of benefit to the agency. LaRC has much more space in technical facilities than it has in office facilities, with the current replacement value (CRV) of the technical facilities dwarfing the CRV of the office facilities (by approximately 30 times). Additionally, the cost for operating the technical facilities is nine times as much as for the office facilities. In the LaRC example, operations- and maintenance-cost avoidance for facilities that have been closed recently is approximately $2.9 million. Annual operations and maintenance cost savings from use of a GIS-based optimization capability associated with facility closures, accrued over multiple years and across multiple centers, would result in hundreds of millions of dollars of benefit to the government. Recently, when Johnson Space Center was evaluating optimization technology for possible investment, it conducted its own return on investment (ROI) analysis. JSC concluded that investment to extend its space utilization with optimization would pay for itself with efficiency benefits if its use of space was compressed enough for one organization not to have to rent outside space during the upcoming mission change. JSC also recognized that effectiveness and less quantifiable benefits would be realized by achieving better synergy within and between their organizations. Space Utilization Optimization J-9780 Esri White Paper 11 Investments in optimization and GIS technology for facilities management in general will propagate to other centers, federal agencies, and possibly nongovernmental facilities. This technology will be especially beneficial to facilities experiencing major organizational changes, consolidation, growth, or downsizing. Modern spatial data management and planning support will enable the facilities embracing the technology to respond more rapidly and efficiently to changing mission goals, budgets, federal requirements (environmental, security), and disaster planning/recovery. Conclusion NASA's human resources and facilities are its greatest strategic assets. The proposed system will optimize the mapping of these assets to achieve center and agency objectives. The system will enable prioritized resource alignment with NASA's strategic objectives. More efficient use of existing and proposed infrastructure will reduce facility-related costs, and more efficient scheduling of resources will minimize delays, thus promoting successful realization of mission goals. LaRC's GIS has produced several major components of an operational software package for space allocation optimization. It provides NASA LaRC and partners with tools for improved decision making, which will result in improvements to LaRC/base operations in cost savings. LaRC's GIS team is currently planning development and integration of many other capabilities. Currently available foundational components include a robust Oracle-based space utilization database with extensive Web interface, GIS-based data management techniques for building interior features that allow overlay of interior details with georeferenced mapping, optimization algorithm for administrative space, and visualization tools such as spatial subdivision and a building/personnel layout diagram. The aforementioned contribute to a discrete optimization toolkit, which produces solutions that minimize the combined costs associated with lack of organizational and functional synergy, inefficient space utilization, and moving costs. XML schema was developed for the space allocation optimization model. The XQuery language was leveraged to produce concise and manageable mapping from the disparate source data into the XML model schema, which dramatically reduced data preparation time. The spatial subdivision diagram visualization tools have also been significantly enhanced, providing decision makers with a comprehensive view of spatially sparse datasets and allowing them to make well-informed choices much more quickly. Work is currently under way to provide a Web interface that will not only display proposed solutions but will also allow manual adjustment of the scenario.

Availability: Out of Stock

Tags :

Billing Information

Order details QTY Total

Product Name 5b6b383e26e7b705484cdf5d

Rs   2100

Note: Delivery usually takes 2.5 days, depending on availability and your location.
Order-Success icon not found

Your order has been placed successfully

Your order has been successfully placed. After it is reviewed, an email with the shipment details will be sent to the email ID you provided.
order-failure icon not found

Your order could not be placed

Your order not be placed. Please try again at a later time.