WHITE PAPER Title: Scalable End-to-End Logistics Simulation (SEELS )

Transcription

1 WHITE PAPER Title: Scalable End-to-End Logistics Simulation (SEELS ) Prepared by 1040 University Boulevard, Suite 100 Portsmouth, VA Phone: (757) Fax: (757) Administrative Point of Contact: Ashley Medford, (757) Authorized Signatory: William Younger President Proprietary (For Evaluation Purposes Only) Page 1 of 9

2 Scalable End-to-End Logistics Simulation (SEELS ) architecture supports the modeling of complex logistical networks for commercial, military, and government customers for use in planning, analysis, and experimentation. For every customer, SEELS models their unique end-to-end logistical network and functions as a decision-support tool to quickly analyze different what-if scenarios. SEELS is a scalable, nodal, and programmable architecture that allows a user to: Define the item (cargo/bulk/person) entity level used in the simulation Scale level of fidelity (as required) to model a node Define the handling processes for a node SEELS can model item flow through any cargo terminal or logistical node, either military or commercial. The definition of a node is limited only by the user s requirements; thus a node may be an entire seaport or a single fueling point along a highway. A set of nodes can then be linked via different transportation modes to model a complete, multimodal end-to-end logistical operation. Further, separate logistical operations can be linked to form an integrated, highly scalable, network-of-networks model. SEELS applies a single-model solution to the end-to-end logistics problem, as a viable alternative to the complexity of a federated solution. Once a logistical plan is modeled within SEELS, users may run extremely high-speed simulations. SEELS executes the movement of hundreds of thousands of items in only a few minutes, allowing users to analyze different whatif scenarios by running repeated simulations of logistical plans applying different assumptions or modifications. SEELS reports the results of these simulations in user selected criteria, such as throughput, wait times, time to execute, arrival profiles, or costs. For any logistical network of interest, SEELS allows: Planning a complete logistics network: Define a local, regional, or global network, with varying levels of fidelity and the ability to scale up or down Define an unlimited variety of logistical nodes:» Warehouses, seaports, airports, distribution centers, fuel points, etc.» Select from a catalogue of prefabricated nodes» Define custom or experimental nodes Define and modify the processes that occur within nodes Link nodes with a variety of transportation links:» Road, rail, air, sea, pipeline, etc.» Define custom links Model people, containers, medicinal equipment, or any variety of material as cargo items Testing a logistics network: Test the network throughput, down to individual cargo item Test for a variety of metrics: time, flow, cost, etc. View material and transportation status across the network Apply natural and man-made disruptions to nodes and links» Study the potential impact on operations» Study proactive strategies to prevent disruptions» Study reactive strategies to minimize impact of disruptions Change the network to analyze different courses of action Experiment with new or conceptual logistical means, processes, or organization to analyze its potential impact Proprietary (For Evaluation Purposes Only) Page 2 of 9

3 Features of SEELS include: NODAL CARGO TERMINAL ARCHITECTURE SEELS possesses the flexibility to model any cargo terminal or logistical node (seaport, airport, rail yard, inland port, distribution center, etc.), whether military or commercial. It is developed for configurability to allow a user to easily model new node types, new scenarios, new processes, and new technologies without requiring software code changes. It enables the specification of different configurations of nodes and the linkages to connect these nodes (even those with different transportation modes) within a single integrated scenario, thereby modeling a complete logistical network. USER-DEFINED PROCESSES SEELS provides the Programmable Process Flow Network (PPFN) language, allowing a user to define, edit, and modify customized processes for each logistical node. PPFN is a process flow programming language that describes the process activities at a cargo terminal. PPFN provides the user with the ability to modify processes independent of the simulation and provide it as an input to the simulation at run-time in the same way other data is input. PPFN provides an intuitive approach for Subject Matter Experts (SMEs) to capture processes within a cargo terminal, by providing a process-oriented interface (in terms of flowcharts) for a discrete-event simulation, with processes being defined as sequences of activities and decisions. This makes SEELS more responsive to operational changes by allowing the user to make rapid changes to the process flows to reflect current operational and analytical requirements. MULTIMODAL TRANSPORTATION MODELING ABILITY SEELS supports multimodal transportation modeling including sea, air, terrestrial, and rail modes of transportation. SEELS supports modeling existing transports (ships, aircraft, trucks, trains, etc.) by accessing their characteristics and capabilities to carry cargo from relevant databases. SEELS also supports experimentation with future transportation assets, by allowing a user to define proposed characteristics and capabilities of the individual asset. USER-DEFINED ENTITY-LEVEL FIDELITY SEELS supports modeling cargo items at any level of detail desired by a user. For example, a user can model cargo at the aggregate STON level, at the individual container level, or at the individual container item level of detail. This allows each application developed using SEELS to focus its analysis at the appropriate cargo level. This allows SEELS to model any type of cargo flow through any type of cargo terminal, including people, medicinal equipment, or commercial containers. PROGRAMMATIC EVENT SPECIFICATION SEELS supports a programmatic event capability, allowing a user or external disruption models to program unusual events to occur at specific simulation times. These events modify normal operations by changing the available resources, capabilities of infrastructure and processing and transit times (specified as simulation input) within a node during simulation run-time. For example, an analyst can program a berth within a seaport to be disabled between days 10-17, in response to a chemical incident at the berth, and then study the results of disabling this berth on the entire operations within the seaport. HIGH-PERFORMANCE SIMULATION FOR RAPID ANALYSES SEELS is implemented using platform-independent C++ source code and utilizes a sequential implementation of DIESEL, a high-performance simulation executive model as its core simulation engine (Mathew 2007). Logistical applications built using the SEELS architecture are capable of very fast simulation execution times. For example, Port Analysis Simulation (PAS) (developed for VPA) simulates the flow of about 45,000 pieces of cargo (executing about 1.6 million events) in under 25 seconds and the flow of over a million pieces of cargo in under 45 minutes. Applications developed using the SEELS architecture can perform rapid analyses at the tactical, operational, and strategic levels. Proprietary (For Evaluation Purposes Only) Page 3 of 9

4 EXAMPLE APPLICATION USING SEELS PORT ANALYSIS SIMULATION (PAS) The Port Analysis Simulation (PAS) models the complete multimodal logistical network for the Virginia Port Authority (VPA) and supports detailed what-if analyses from an operations and security perspective. PAS, developed by using the SEELS architecture, models commercial operations (infrastructure, resources, business rules, cargo flow, etc.) within all of VPA terminals comprising the Port of Hampton Roads, including: APM Portsmouth Marine Terminal (APMT) Norfolk International Terminal (NIT) Newport News Marine Terminal (NNMT) Portsmouth Marine Terminal (PMT) Virginia Inland Port (VIP) PAS connects together relevant cargo terminals (seaside and inland ports, rail yards, etc.) and the transportation infrastructure between these terminals, thereby presenting a system-level view of cargo operations for the entire Hampton Roads region. PAS enables operational planners to model and analyze the current state of VPA operations using: Infrastructure data and graphical views populated from GIS databases Resource data populated from VPA asset databases and public data sources Processes derived from business rules specified by VPA Cargo and transport flow scenarios populated from data provided by VPA Cost data for infrastructure, resources and operations populated from data provided by VPA PAS enables VPA to model their current operations across all of its terminals to study the normal state of operations and validate current cargo throughput. PAS is capable of determining cargo throughput capability using current assets, resources, and business processes. It tracks utilization of critical resources and identifies potential bottlenecks and limiting resources to cargo movement through the network. PAS also supports what-if analysis on the impact of adding new infrastructure (new berths, or increases in staging capacity) and/or new resources (straddle carriers, container cranes, etc.). VPA can use this analysis to determine the impact that the increased capability has on performance metrics such as, cargo throughput and truck turnaround times. The analysis can also be used to calculate Return on Investment (ROI) metrics before significant investment decisions are made. VPA can also model proposed changes to its business processes and determine the impact (positive or negative) on the same performance metrics, before actually implementing the changes. VPA can also use PAS to model the planned Craney Island Terminal scheduled to be operational after 2020, using its proposed resources and infrastructure to simulate the flow of projected cargo through the terminal to validate current assumptions and identify future requirements. PAS also supports security analyses and assists VPA operational planners to develop contingency plans for various disruptive scenarios and plan for new security measures. The first type of security analysis is a proactive or preventive approach, where areas and operations within the ports and the transportation infrastructure are analyzed to identify their impact on operations should they be disrupted. Planners then develop efforts to protect those capabilities using different methods such as new security measures, increased container and truck inspections, etc. PAS allows planners to insert proposed inspection stations anywhere within the port facilities (between berth and staging, or between staging and the gates) and determine the slowdown impact that it has on cargo flow. Planners can also use PAS to determine capability requirements for inspection stations (e.g. 40 inspections per hour) to minimize the impact on cargo flow. Proprietary (For Evaluation Purposes Only) Page 4 of 9

5 The second type of security analysis is to analyze responses to disruptive incidents if they do occur. PAS models a disruption in the operations of a terminal by appropriately reducing the allocation and number of resources, available infrastructure, and processing capabilities of that particular terminal, either permanently or for a specific duration of time within the simulation. The impact of response strategies to alleviate slowdown in cargo flow can then be analyzed including cleanup/recovery procedures and rerouting of cargo to maintain operational effectiveness. This demonstrates the effect on a terminal and the potential ripple effect on other terminals within the region as the logistical network adapt to sustain cargo flow. PAS can also be integrated with other models (traffic models, weather models, etc.) to form a complete federation that includes all ports and transportation networks in the region and accounts for the man-made and natural variables that can impact port operations. It can also be interfaced with validated disruption models that can automatically inject stimuli to modify the behavior of the simulation. PAS GUI: Solution Designer Data Sources GUI: Data Analyzer Charts & Graphs VPA CONOPS Transport & Cargo Databases GIS & VPA Asset Databases Disruption Model(s) Financial Factors Rate Data Performance Analysis PPFN Language Transport & Cargo Node Layout Disruptive Event Cost Bottleneck Analysis Cost Analysis Input Data Output Data SEELS Simulation Core PPFN Virtual Machine Nodal Cargo Terminal Architecture DIESEL Proprietary (For Evaluation Purposes Only) Page 5 of 9

6 EXAMPLE APPLICATION USING SEELS PORTSIM 6.5 PORTSIM is a discrete-event simulation that facilitates the analysis of movements of military unit equipment through worldwide seaports and allows for detailed infrastructure analysis. PORTSIM simulates the use of seaports within the defense transportation system. It determines port throughput capability given explicit assumptions on assets, resources, and scenarios. It tracks utilization of critical resources as well as potential bottlenecks and limiting resources to movement through the seaport. PORTSIM also supports analyses by the Surface Deployment and Distribution Command Transportation Engineering Agency (SDDCTEA) and the Joint Distribution Process Analysis Center (JDPAC), both USTRANSCOM components. PORTSIM assists SDDCTEA and JDPAC analysts in comparing and selecting ports for movement of military cargo based on their capabilities and/or refining their specific strategies for movement of a particular set of materiel through a selected port. PORTSIM 6.5, developed by using the SEELS architecture, integrates the Seaport of Debarkation (SPOD) and Seaport of Embarkation (SPOE) components into a single integrated model, models the resources and infrastructure within a seaport (including terminals and operational areas), and supports the configuration of an individual seaport according to its characteristics. PORTSIM 6.5 models competition for resources (cranes, drivers, etc.) and infrastructure (berth space, staging space, etc.) by cargo and transports within a seaport. It supports the concurrent SPOD and SPOE flow through a single seaport. This makes PORTSIM 6.5 an important tool in the analysis of the flow of military cargo, aids the decision-making process increasing both the efficiency and throughput of any military deployment or redeployment operation. PORTSIM 6.5 provides SDDCTEA with the ability to simultaneously study the SPOD and SPOE modes of operation using a single tool, with the same level of fidelity and identical input and output data formats (Mathew, Mastaglio, and Leathrum 2009). SDDCTEA Transports Database SPOE & SPOD Process Flows PORTSIM 6.5 SDDCTEA GIS Database PREPROCESSOR PORTSIM 6.5 Proprietary (For Evaluation Purposes Only) Page 6 of 9

7 EXAMPLE STUDY USING SEELS Intratheater Sealift Operations Intratheater sealift operations support the movement of military cargo into areas where there is insufficient access to world-class seaports to support a standard military deployment. The operation utilizes a large, world-class seaport of debarkation, located a safe distance from the region of conflict, as a Sea-based Intermediate Staging Base (SISB). At the SISB, cargo is transferred from large cargo transport ships to smaller, faster vessels called Theater Support Vessels (TSVs) that are capable of utilizing small or reduced capacity ports in the region. The SEELS architecture is used in a prototype for modeling intratheater sealift operations, where a set of ports in Hawaii, Japan, and Korea are modeled. The fictitious scenario requires the deployment of a US mechanized division from its home base in Hawaii to south-central South Korea. The division is transported by large cargo transport ships from Hawaii to the port of Naha, Japan (acting as the SISB) where the division is loaded onto TSVs and are transported to four Seaport of Debarkations (SPODs) having austere capabilities located on the southern and eastern coasts of South Korea for rapid deployment into the theater of operations (Leathrum, Mielke, Mazumdar, Mathew, Manepalli, Pillai, and Malladi 2004; Mathew, Leathrum, Mazumdar, Frith, and Joines 2005). The scenario is shown below. Convoys Pearl Harbor FISC Area Terminal West Loch Terminal (SPOE) Ships Theater Military Depot (Installation) Naha, Japan (SISB) TSVs Convoys Chinhae, South Korea Masan, South Korea Pusan, South Korea Pohang, South Korea (RPODs) The prototype is used to: Identify number of TSVs required to successfully execute an Intratheater Sealift operation Identify future TSV capabilities (capacity, loading/offloading rate, and speed/range) Calculate ideal proximity between a SISB and RPODs Recommend force reconfiguration to be lighter, more mobile, and self-deployable Proprietary (For Evaluation Purposes Only) Page 7 of 9

8 EXAMPLE STUDY USING SEELS Modeling End-to-End Logistics The SEELS architecture can model the end-to-end flow of cargo at an entity level in the form of a nodal network. The architecture deals with resource and infrastructure allocations and competition within each node in the network, as well as between the nodes, to identify conflicts that may arise. It can model any scenario where largescale logistical support is required such as military deployment/redeployment scenarios (using the defense transportation system), humanitarian aid scenarios, commercial distribution networks, etc. A prototype for modeling multimodal end-to-end logistics from multiple CONUS-based installations to a set of destinations within a theater of operations located anywhere in the world is described in (Mathew, Leathrum, Mazumdar, Frith, and Joines 2005). was invited to the 5 th Annual Capitol Hill Modeling and Simulation Expo in June 2010, where we demonstrated the use of the SEELS architecture in a disaster relief scenario deploying humanitarian aid, personnel, and cargo from CONUS to Grenada in response to a hurricane. In the scenario, logistical planners are required to quickly create a logistics plan to support a complex aid flow and wargame different responses using a variety of nodes (seaports, airports, distribution centers, etc.) and a variety of transportation means (sea, air, convoys, rail, etc.). The scenario is shown below. Warehouse Convoys SPOEs Hampton Roads Newport News Marine Terminal Portsmouth Marine Terminal Savannah Ocean Terminal SPOD St. George, Grenada St. George Terminal A St. George Terminal B Convoys with Medical Supplies Ships UAVs with Relief Supplies Aircraft APOEs Charleston AFB, SC Homestead Air Reserve Base, FL Helos with APOD Personnel Maurice Bishop Intl. Airport Military Base Rail Proprietary (For Evaluation Purposes Only) Page 8 of 9

9 REFERENCES Mathew, R. "The Distributed Independent-Platform Event-Driven Simulation Engine Library (DIESEL)." PhD Dissertation, Doctor of Philosophy in Electrical and Computer Engineering, Department of Electrical and Computer Engineering, Old Dominion University, May Mathew, R., and J.F. Leathrum, Jr. "Programmable Process Language for Modeling Cargo Logistics." Proceedings of MODSIM World 2008, September , Virginia Beach, VA, USA. Mathew, R., T. Mastaglio, and J.F. Leathrum, Jr. "PORTSIM 6.0: A Port Simulation Modeling Multiple Modes Of Operation." Proceedings of the International Conference on Harbor, Maritime & Multimodal Logistics Modelling and Simulation (HMS 2009), pp , September 23 25, 2009, Puerto de La Cruz, Spain. Mathew, R., J.F. Leathrum Jr., S. Mazumdar, T. Frith, and J. Joines. "An Object-Oriented Architecture for the Simulation of Networks of Cargo Terminal Operations." Journal of Defense Modeling and Simulation (JDMS), vol. 2, no. 2, , April Leathrum, J.F., Jr., R.R. Mielke, S. Mazumdar, R. Mathew, Y. Manepalli, V. Pillai, and R.N. Malladi. "A Simulation Architecture to Support Intratheater Sealift Operations." Mathematical and Computer Modelling, vol. 39, no. 6-8, , May Leathrum, J.F., Jr., R. Mathew, and T. Mastaglio. "Modeling & Simulation Techniques for Maritime Security." Proceedings of IEEE International Conference on Technologies for Homeland Security (HST 2009), Proprietary (For Evaluation Purposes Only) Page 9 of 9