Attachment 1 REBASE Overview_Final.docx

Attachment 1

Redefining the Basic Analytical and Simulation Environment (REBASE) Overview

April 2, 2020

Background and Purpose Background The Aerospace Vehicle Technology Assessment and Simulation (AVTAS) branch in the Air Force Research Laboratory (AFRL) currently develops, maintains and distributes the Advanced Framework for Simulation, Integration, and Modeling (AFSIM), a modular, object-oriented Multi-Domain Operations (MDO) Modeling, Simulation, & Analysis (MS&A) framework. AFRL intends for AFSIM to support multi-domain, multi-resolution MS&A activities ranging from engineering to mission-level into campaign. Beyond the framework itself, AFRL develops and maintains an extensive AFSIM-compatible toolset including applications to ease scenario and model creation, improve results visualization, and enhance operator-in/on-the-loop interactions.

AFSIM and its ecosystem of tools have evolved significantly over the past 15 years. Created out of frustration with legacy Air Force MS&A tools, AFSIM originally provided a flexible solution to study advanced technology concepts within the air domain at the mission-level. AFRL has since expanded AFSIM’s capabilities to partially address all warfighting domains (air, land, sea, space, cyber) as well as spectrum of simulation needs from the engineering-level to the campaign-level. The AFSIM community, including government and industry users across the DoD and Intelligence Community (IC), have encouraged these investments by continuing to supply AFRL with new and challenging requirements. Meanwhile, growing interest in AFSIM has postured the framework to become a common tool in the DoD’s MS&A toolbox. To date, AFRL has distributed AFSIM to over 300 government and industry organizations and trained over 1500 AFSIM users and developers.

Challenge and Opportunity The DoD’s aggressive pivot towards MDO requires a force that can achieve dominance in one warfighting domain or many, blending few capabilities or many, to achieve overwhelming freedom of action and superiority against U.S. adversaries, particularly near-peer competitors (Air Force Chief of Staff Gen. David L. Goldfein – July 2019). Poised to deliver material solutions that realize this operational reality, the DoD science and technology (S&T) enterprise requires new model-based methods and tools to study both the problem and solution spaces associated with MDO. Likewise, MDO represents a total-force challenge that requires full reach-back into the U.S. defense innovation base.

While AFSIM has gradually improved its ability to represent concepts across warfighting domains, it requires an equally aggressive investment to become a truly formidable MDO tool in the hands of scientists, engineers, and analysts. Certainly, AFSIM’s open business model and flexible Information Transfer Agreement (ITA) have motivated its ascent as a common tool. However, its “out-of-the-box” or “boxset” capabilities have not yet reduced start-up costs associated with MDO analysis. Furthermore, while AFRL has advanced AFSIM’s commoditization and maturation, the framework has yet to reach its full potential as a stable platform supporting the rapid and evolutionary exchange of multi-domain concepts and solutions among its users.

While many government organizations use AFSIM or fund AFSIM-related work, this effort intends to address its mainstream development to address the MDO challenge.

Vision and Scope Anticipated Period of Performance and Budget The shared ceiling for the Redefining the Basic Analytical and Simulation Environment (REBASE) Program will be $48 million over a four-year period of performance.

Vision Under REBASE, AFRL intends to use AFSIM to advance the state-of-the-art in DoD MS&A and respond to the evolving needs of the defense innovation base as it accelerates the research, discovery, development, and transition of novel multi-domain warfighting capabilities.

To promote this vision, AFRL has established three broad objectives:

· Objective-1: Accelerate Multi-Domain Operations MS&A

· Objective-2: Evolve Foundational MS&A Frameworks and Tools

· Objective-3: Deliver MS&A Capabilities at the Speed of Relevance For each objective depicted in Figure 1, AFRL has also planned several initiatives and elaborated focus areas to both refine scope and convey explicit priorities and needs.

Figure 1: REBASE Objectives Together, these objectives and initiatives trace back to the challenges and opportunities enumerated in Section 1.2. These initiatives generally respond to the need to expand multi-domain “boxset" capabilities, the need to address the full complexity of the multi-domain context, and the need to rapidly evolve software products through Agile principles and practices characteristic of a modern digital enterprise.

Scope AFRL’s objectives, introduced in Section 2.2 and detailed through initiatives in Section 3 should culminate in a balanced product that can more flexibly iterate between analytical concerns, from engineering to campaign levels.

For many decades, the military operations analysis and modeling communities have added a number of tools to the DoD “toolbox” to address specific questions that arise during technology development and acquisition. The familiar MS&A Pyramid depicts five distinct levels of modeling resolution/detail appropriate to answer such questions. Table 1 characterizes each level through typical attributes and uses. Over time, modelers and analysts have had to learn to use and integrate an ensemble of simulation tools, each serving to answer a narrow set of questions within a narrow set of warfighting domains. While AFSIM should not (cannot) subsume all of these specialized capabilities under the umbrella of a single multi-domain framework, it should help end-users more flexibly and efficiently iterate between analytical concerns. To fulfill this need, AFSIM requires balanced investment where it can well serve its “home ground” for engagement to mission-level modeling while reducing time and cost for carrying analytical investigations to the engineering and campaign levels.

Table 1: Levels of Analysis

Individual components, interactions, physics Individual entities;

Detailed subsystems Individual entities;

Moderately detailed subsystems Many individual entities;

Some aggregated entities;

Detailed IADS Highly aggregated entities

Typical Metrics Prob. of kill;

Vulnerable area Miss distance;

Prob. of kill;

Acquisition range Loss exchange ratio;

Survival/Loss rate;

Missile performance Survival/Loss rate;

Prob. of engagement;

Missile shots/kills Expeditionary forces flow;

Target drawdown;

Target kills;

Campaign cost

Typical Uses Design trades;

Performance trades;

Requirements development;

Test;

Tech Investment Analysis of Alternatives (AoAs);

Design trades;

Performance trades;

Requirements development;

Tactics evaluation;

Test;

Tech Investment AoAs;

Design trades;

Performance trades;

Requirements development;

Tactics evaluation;

Test;

Tech Investment

AoAs;

Requirements development;

Concept of Operations (CONOPS);

Operational Concepts;

Test;

Tech Investment

AoAs;

Force structure design;

Requirements development;

War gaming;

Sustainment Mobility & logistics;

Military worth/utility;

CONOPS;

Tech Investment

Business Strategy Under REBASE, AFRL intends to prototype an agile governance and competitive acquisition approach that quickly responds to evolving multi-domain requirements. This approach should also help catalyze AFSIM’s growing user community into a robust supply chain for novel model-based products and tools.

To describe the enterprise under REBASE, AFRL will refer to the following stakeholder roles:

1. The AFSIM Program Management Team (PMT): Government stewards who ultimately define AFSIM’s business and technical strategy and exercise governance authority over all community-driven aspects of the product

2. Framework Developers: Principal members of the AFSIM community advancing the core AFSIM framework and applications that support both end-users and other capability providers

3. Capability Providers: Members of the AFSIM community that build upon or extend the core AFSIM framework to provide value to end-users in the form of new models, scenarios, plugins, applications, and associated support products

4. End-Users: Members of the AFSIM community (scientists, engineers, and analysts) who ultimately use AFSIM to discover, develop, and transition novel multi-domain warfighting capabilities Each role is pivotal to the successful outcome of each objective and the continuous and incremental delivery of value to end-users throughout the period of performance.

Acquisition Approach AFRL’s acquisition approach will favor shorter (6-12 month), well-scoped, development task orders coinciding with a continuous AFSIM release cycle. AFRL intends to maximize opportunity for multiple Capability and Framework Providers to contribute to the product. Figure 2 depicts this approach in a notional form (number, frequency, scope, and duration of task orders subject to change).

Figure 2: Acquisition Approach The AFSIM PMT will assume a Product Owner role by continuously managing, prioritizing, and refining both government and end-user needs in a Product Backlog. Upon verifying that these needs fall within the scope of REBASE, The AFSIM PMT will plan to address these needs across a series of task order solicitations. AFRL has identified at least three themes, traceable to one or more REBASE objectives, to help inform the scope of each series of task orders.

Expansion Task Order Series AFRL intends to solicit a series of task orders that address gaps in AFSIM’s “boxset” capabilities delivered to end-users in the form of AFSIM-compatible models, scenarios, and plugins. AFRL anticipates that a majority of these task orders will fall under the scope of REBASE Objective-1 and will need to tap into a broad pool of technical and operational expertise from across the AFSIM community. AFRL anticipates that this series may introduce both single and multiple award opportunities.

Experimentation Task Order Series To manage technical risk and ensure best of breed solutions eventually make their way into the official AFSIM release, AFRL intends to solicit periodic task orders to charter prototypes, experiments, and short studies. AFRL anticipates that a majority of these task orders will fall under the scope of REBASE Objectives 1 and 2 and may help to expand the pool of contributors intimately familiar with AFSIM’s architecture and internals. AFRL anticipates that this series would predominantly favor multiple award opportunities.

Integration Task Order Series AFRL’s ability to broadly solicit and acquire new AFSIM-compatible models, scenarios, and plugins depends upon a stable architecture and technical foundation. To provide that foundation, AFRL intends to solicit a series of task orders that advance the capabilities of the core framework for a growing pool of potential contributors. This series will also establish enduring processes and tooling necessary to accelerate the delivery of new features and capabilities developed across multiple task orders. AFRL anticipates that a majority of these task orders will fall under the scope of REBASE Objectives 2 and 3 and predominantly favor single award opportunities fulfilling a critical integrator role.

Tiered Ordering Structure In adherence to a modular and streamlined contracting approach, AFRL will adopt a tiered ordering structure. Throughout the period of performance of this effort, AFRL will solicit task orders that fall into one of three categories described in Table 2.

Table 2: Tiered Ordering Structure

· Duration 3-6 months

· Shorter proposals

· Duration 6-9 months

· Duration exceeding 9 months

Coordination Strategy AFRL’s business strategy requires an integrated, “whole team” approach that realizes a shared enterprise among multiple parties – the AFSIM PMT, Capability Providers, Framework Developers, and End-users. While AFRL can assume a more active role as Product Owner, it cannot assume technical coordination and integration responsibilities. For this reason, AFRL has identified several principal technical roles that fall within the scope of the Integration Task Order Series. As described in Table 3, these roles include, but are not limited to, the Project Manager, Principal Architect, and Release Engineer. While offerors may define additional roles and should dynamically adjust their team composition and structure to meet requirements, these principal roles will persist between task orders to ensure continuity of coordination.

Table 3: Principal Roles

Program Manager (Government)

Business Manager (Government)

Contracting Officer (Government)

Product Owner (Government)

Lead Software Engineer (Government)

Project Manager (Contractor)

Principal Architect (Contractor)

Release Engineer (Contractor)

Project Manager Role The Project Manager is the primary point of contact (POC) responsible for coordination among the offeror’s team and members of the AFSIM PMT.

Responsibilities should include:

· Communicating with the Program Manager, Business Manager, and Contracting Officer to enable timely resolution of programmatic issues and blockers

· Identifying and fulfilling project staffing requirements

· Coordinating Agile “ceremonies” such as release and iteration/cycle/sprint planning meetings

· Facilitating project reporting and metrics in line with Agile planning, estimation, and control

· Coordinating with the Product Owner to develop and groom the Product Backlog Principal Architect Role The Principal Architect is responsible for planning and executing architectural investments (Agile terms: managing “technical debt” and extending the “architectural runway”) in support of planned features and capabilities in close coordination with the Product Owner and Lead Software Engineer.

Responsibilities should include:

· Coordinating with the Product Owner and associated stakeholders (end-users, sponsors, suppliers) to define and communicate key architectural qualities and intent guiding the evolution of the AFSIM framework

· Translating AFRL’s business and technical strategy to the development team and associated suppliers

· Overseeing the execution of technical work to enforce key architectural qualities (e.g., suitability, compatibility, usability, flexibility, performance) through design and implementation guidelines

· Coordinating with the Lead Software Engineer to foster a Continuous Incubation and Exploration process that drives investigation of technical alternatives through experimentation (Agile term: "technical spikes”)

· Participating in Agile “ceremonies” that implicate technical planning, requirements definition, and high-level solution design and evolution Release Engineer Role The Release Engineer is responsible for overseeing the development, security, and operations (DevSecOps) Continuous Delivery pipelines and associated quality assurance activities in close coordination with the Lead Software Engineer. This role has assigned cybersecurity functions and must maintain Information Assurance System Architect and Engineer (IASAE) Level-II certification in accordance with AFMAN-17-1303 and DoD 8570.01-M.

Responsibilities should include:

· Coordinating all development, testing, distribution, deployment and end-user support/operations concerns across the offeror’s team

· Participating in Agile “ceremonies” that implicate release planning, configuration control, build automation, test automation, deployment, development and production environments, and associated cybersecurity controls

· Instrumenting Continuous Delivery pipelines to provide relevant product and process metrics that support agile data-driven decision making

· Coordinating with the Lead Software Engineer to shepherd AFSIM through DoD cybersecurity and continuous assessment processes

· Overseeing all platform engineering activities necessary to support the continued development and evolution of AFSIM on Government-managed infrastructure

· Driving maturation of the M&S-as-a-Service concept and supporting cloud-native design, deployment, and operations as described in Objective-2 (Section 3.2.3.5) Shared Infrastructure and Platforms Policy AFRL is not only interested in building new AFSIM capabilities, it is also interested in building the enterprise required to evolve AFSIM for the future. That is, AFRL views investment in the Software Factory equal in value to the software that it produces. Consequently, AFRL does not believe that the general processes and tooling required to produce or evolve AFSIM over its lifecycle should become unique to any particular vendor. AFRL intends to support this goal by migrating as many AFSIM development and support activities to Government-managed infrastructure and platforms as practically possible.

Specific Agile and DevSecOps practices promoted under this effort assume the availability of a development platform with a suitable technology stack. While in the past, AFRL could not reliably furnish these capabilities, recent enterprise-level investments and initiatives have drastically changed this calculus. Offerors should now maximize use of Government-managed infrastructure and seek authorization from the Government before deviating from a full migration goal.

Under this effort, AFRL intends to promote the use of the Defense Intelligence Information Enterprise (DI2E) provided technology stack (DevTools) as a shared development platform for all objectives (https://www.di2e.net). While AFRL believes that DI2E DevTools provides a sufficient set of capabilities to accomplish its objectives, it also seeks to leverage emerging investments from the DoD DevSecOps Initiative such as Cloud One and Platform One.

AFRL recognizes several risks to this strategy, particularly in the case where multiple contractor teams are conducting parallel development activities on a shared platform for a common product. To mitigate these risks, AFRL desires to collaborate with offerors who can fully embody principle technical roles such as the Principal Architect and Release Engineer. AFRL has vested these roles with an appropriate vantage point and technical authority to avoid potential issues related to scale and complexity.

Data Rights Strategy Pursuant to its business strategy, AFRL intends to acquire “Unlimited Rights,” as that term is defined under 48 C.F.R. § 252.227-7013(a)(16) and 48 C.F.R. § 252.227-7014(a)(16), to any software, documentation, and tooling developed under this effort. Software and documentation includes, but is not limited to source code, data, documents, graphics, plans, schedules, schemas, metadata, architecture designs, training/support data, and the like; and all new tooling, infrastructure as code, script configuration management, or any other end-items necessary for the community-wide development, maturation, and unimpeded distribution of AFSIM under the liberal terms and conditions of the AFSIM ITA.

Objectives Accelerate Multi-Domain Operations MS&A Objective-1 directly supports the vision of this effort by developing the “boxset" capabilities need to reduce start-up costs associated with MDO MS&A.

Section 3.1.1 describes AFRL’s motivation and purpose for defining this objective, Section 3.1.2 describes its specific scope of focus, while Section 3.1.3 enumerates five key initiatives within that scope.

Purpose AFRL intends for AFSIM to support comprehensive MDO MS&A for the rapid study of emerging military technology concepts. Rich multi-domain representations help technologists, engineers, and analysts better assess military utility, evaluate architectural tradeoffs, and make informed engineering decisions. Few organizations maintain a “deep bench” of subject matter experts across disciplines who can develop these representations from scratch. For smaller organizations or shorter studies, this is often a non-starter. Developing, verifying, and (depending on the use) validating models and scenarios demands time, cost, talent, and especially, data. Ultimately, project teams lose precious schedule time while training personnel on tools, wrangling data sets, and tuning models for study purposes. AFRL recognizes these challenges and intends to lower these barriers by expanding AFSIM’s “boxset” MDO capabilities. Characteristics of an ideal MDO baseline include multi-resolution model and scenario representations at various degrees of fidelity, pedigreed data with well-documented traceability, support for mixed live, virtual, and constructive simulation, and a well-curated library of red and blue systems. When provided such a research accelerant, end-users can more efficiently apply their specific domain expertise to the essential analytical questions.

Scope AFRL intends to significantly expand and improve AFSIM’s “out-of-the-box" model, scenario, and capability set. Five Government identified initiatives (Table 4) describe the scope of this investment.

Table 4: Objective-1 Scope

Initiatives Enhance Air Combat Operations Capabilities Beginning with its legacy heritage and continued through Air Force sponsorship, AFSIM has incrementally improved its representation of air assets and their agency through combat behaviors and tactics. Investments to-date have helped reduce the time required to develop new aircraft and missile models sensitive to physics-based effects (e.g., aerodynamics, aircraft handling characteristics, propulsion), imbue those models with script-based perception and cognition, and understand their performance through visualization and analysis tools. Continued expansion and maturation of these investments will ensure that AFSIM can meaningfully contribute to the development next-generation air dominance capabilities. This initiative focuses on building up AFSIM’s catalog of air asset models, expanding its library of modular air combat behaviors and tactics, and improving the methods and tools by which end-users can generate, compose, and explore those behaviors.

Enhance Space Operations Capabilities Spaceborne capabilities have become an integral part of modern terrestrial warfare. As a result, mission and campaign-level simulation requires increasingly more detail in the depiction of space operations. The ubiquity of space effects on all warfighting domains drives a real need for “boxset” space capabilities to accommodate multiple levels of fidelity and degrees of configurability. While some end-users need authoritative, “plug-n-play” model assets to build a plausible space context, others require more flexibility to explore system-of-system tradespaces characterized by the performance of these spaceborne assets. Over the past several years, AFRL has invested significantly in improving AFSIM’s representation of the space domain. Capabilities range from space weather modeling, to common propagator models, to visual tools that aid in the rapid creation of satellite constellations. This initiative intends to expand AFSIM’s space operations capabilities in the form of models and supporting tools while balancing needs across the user community.

Enhance Cyber and EW Operations Capabilities Success in modern warfare depends upon the control and flow of information, the ability to disrupt or influence an adversary’s decision-making process, and the capacity to defend against the same. In order to maintain strategic advantage, the DoD continues to emphasize the criticality of information operations characterized by dominance of the entire electromagnetic spectrum and vital communications, command, and control networks and systems (JP 3-51). The capacity of U.S. adversaries to sense or affect the electromagnetic spectrum and vital defense systems presents a particular challenge for scientists, engineers, and analysts developing future weapon systems. Likewise, new materiel solutions (including both red and blue systems) often present new attack surfaces, exploited exclusively through the electromagnetic/information environment. AFSIM has traditionally represented the aggregate mission-level effects of electromagnetic and information operations on modeled systems and networks. While this level of fidelity has satisfied basic operational effectiveness studies, it strains to support studies that require more spatially and temporally accurate detail about the electromagnetic/information environment and interactions therein. This initiative intends to improve the capacity of “boxset” models to participate in these contested environments while providing end-users with a powerful set of analytical tools to master its effects.

Expand Threat Representations The DoD has steadily increased its reliance on MS&A to make significant S&T investment and acquisition decisions. With limited resources and significant opportunity cost at stake, decision makers have likewise raised the bar for technologists and analysts to defend the credibility of their analyses. A successful defense generally begins with the use of traceable, pedigreed, validated, or otherwise authoritative, intelligence-derived models and scenarios. The AFSIM “boxset” should provide both end-users and decision makers alike with the evidence necessary to substantiate critical claims and arguments expressed through simulation. Ideally, system owners and intelligence agencies would provide end-users with comprehensive data on both foreign and friendly system-of-systems representations. Unfortunately, the DoD has yet to establish an accessible model clearinghouse inclusive of all warfighting domains and concerns. Where data exists, end-users must often make informed inferences or “port” existing models between frameworks and simulation environments. These translation steps introduce new sources of error and analytical doubt. To improve analytical credibility, AFRL intends to focus on the continuous and incremental expansion of authoritative, intelligence-derived threat representations and scenarios in AFSIM. These threat representations should come from supported IC model product lines, ensuring that the IC remain the recognized source for authoritative, intelligence-derived models in AFSIM. Model production should include commensurate investment in the tooling and processes necessary to accelerate threat modeling, integration, and verification in AFSIM.

Under this initiative, the following focus areas prioritize investments that will provide authoritative threat representations in the AFSIM “boxset” across a range of fidelities and domains and improved tooling to support threat modeling and verification.

· Incorporate Pedigreed Threat Models

· Incorporate Pedigreed Threat Scenarios Expand Joint Operations Capabilities AFRL intends for AFSIM to become a commodity tool useful across the entire defense M&SA community. Toward this end, the AFSIM user community has grown significantly over the past seven years and now includes representation across all military services and intelligence production centers. While AFRL does not intend to fashion AFSIM as a replacement for service and domain-specific models, it does intend to improve its representation of joint operations capabilities. More specifically, a meaningful pivot towards all-domain analysis requires balanced attention to the maritime and land operational domains. Historically underrepresented in AFSIM “boxset” scenarios and models, investments in these domains can help “unlock” new capabilities while re-imagining the future of U.S. combat operations.

Under this initiative, the following focus areas prioritize investments in the expansion of joint capability representations with the goal of posturing the framework to address the needs of the whole AFSIM user community.

· Expand Maritime Operations Capabilities

· Expand Land Operations Capabilities Evolve Foundational MS&A Frameworks and Tools Objective-2 directly supports the acceleration of MDO MS&A by anticipating and fulfilling the foundational tooling and framework needs for Objective-1.

Section 3.2.1 describes AFRL’s motivation and purpose for defining this objective, Section 3.2.2 describes its specific scope of focus, while Section 3.2.3 enumerates five key initiatives within that scope.

Purpose AFSIM and its ecosystem of tools must continue to adapt to new drivers and use cases traceable to the unique challenges of studying novel warfighting concepts in a multi-domain context. Propelled by the commoditization of computing and the practical application of artificial intelligence in the commercial sector, warfare is again at the cusp of change. Operations and systems will rapidly evolve in both structural and behavioral dimensions of complexity while levying new requirements against analytical tools to help navigate these decision spaces.

AFRL has uniquely positioned AFSIM to respond to these changing demand signals. Harkening back to its heritage in the Analytic Framework for Network Enabled Systems (AFNES) over a decade ago, AFSIM has provided its users with a malleable medium by which to express new operational and system concepts. While the framework has matured significantly over the years, its component and plugin-based architecture, integrated and extensible suite of analyst tools, and modern development environment represent an investment in and testament to its future evolvability.

AFRL must address the challenge of quickly responding to these change drivers while also providing a stable and suitable platform for its stakeholders. As end-users build new capabilities upon this platform in the form of AFSIM-compatible models, scenarios, and plugins, there is an assumption that user-generated investments can readily travel alongside the AFSIM lifecycle. Likewise, AFRL recognizes that an expanding set of multi-domain capabilities can quickly overwhelm the capacity of any one user or developer to understand, use, or maintain.

Scope AFRL intends to promote five essential qualities (Table 5) to guide the evolution of AFSIM as a foundational tool for MDO MS&A. Under this effort, offerors will realize five Government identified initiatives to drive these qualities into the current design and implementation of AFSIM as well as the architecture for the next major version.

Table 5: Objective-2 Scope

Enrich Multi-Domain Foundations MDO MS&A presents an opportunity to expand beyond traditional platform-centric thinking. Under traditional modeling paradigms, the platform is the fundamental building block of a simulation and specifies both the structural and behavioral aspects of any simulated actor. However, much of the complexity of MDO manifests in the behaviors, interactions, timing, and information flows between actors (platforms) across domains in an integrated battlespace. This initiative revisits core AFSIM framework decisions in light of multi-domain requirements with the intent of enhancing, extending, and re-architecting (where necessary) to provide a complete set of capabilities for multi-domain MS&A.

Under this initiative, the following focus areas prioritize investments that will improve the suitability and long-term viability of AFSIM for future multi-domain technology representations and assessments.

· Enhance Multi-Domain Perception and Battlespace Awareness

· Enhance Multi-Domain Command, Control, and Communications

· Enhance Multi-Domain Maneuver, Fires, and Effects

· Support Complex Environmental Operating Conditions

· Support Limited Campaign and Theater-Level Concepts Join the Digital Ecosystem Digital Engineering is a DoD initiative that intends to refashion the DoD culture and workforce and transform the state of engineering and acquisition through innovative model-based tools and approaches. Consequently, the long-term viability of AFSIM depends upon its ability to fit within a growing ecosystem of specialized engineering and analysis tools. Within this ecosystem, AFSIM would become both a provider and consumer of digital information. To become a “digital native”, AFSIM must provide stable interfaces to support its use alongside a growing suite of community-adopted tools. AFSIM must also provide capabilities to source and use external models or leverage specialized digital services from other providers. These investments will ultimately promote AFSIM as a more useful capability in joint DoD research, experimentation, engineering, and acquisition.

Under this initiative, the following focus areas prioritize investments that will promote “out-of-the-box" compatibility trusted by the community as seamless, reliable, and performant.

· Promote Simulation Interoperability

· Promote Data Interoperability and Model Interchange

· Embrace Emerging Open System Architecture Standards Improve the User Experience The AFSIM user community includes a wide range of domain-specific experience, competency, and skill often characterized using the simplified dichotomy of “analyst” or “developer.” Among “analysts”, this population includes users who desire complete GUI-driven scenario configuration tools to users who desire a comprehensive understanding of the simulated environment. Among “developers”, the population includes users who work within the confines of standard application programming interfaces (APIs) to code contributors who try to enhance all aspects of the framework. Regardless of “home ground” perspective, all AFSIM users require tools to grapple with the exponential growth in complexity introduced with multi-domain system-of-system concepts and interrelated effects chains.

Under this initiative, the following focus areas prioritize investments that will improve the overall user experience and usability of the AFSIM framework and its accompanying tools.

· Improve the Analyst Experience

· Improve the Developer Experience Promote Flexibility In-Use As a framework, AFSIM provides inherent flexibility for end-users do describe an operational scenario of interest and study the mission-level effects of new warfighting concepts virtually and constructively. As the S&T landscape changes rapidly, flexibility also ensures that scientists, engineers, and analysts can continue to lean on familiar tools to solve new challenges. AFSIM’s current flexibility largely follows from its generic component-based architecture, the principal investment over the lifecycle of AFSIM 2.x. However, emerging multi-domain requirements demand additional dimensions of flexibility to include support for modeling and analysis at multiple levels of resolution and classification.

Under this initiative, the following focus areas prioritize investments that will improve the overall flexibility of the AFSIM framework to model concepts at multiple levels of spatial and temporal resolution as well as support analyses that span multiple levels of classification.

· Support Modeling and Analysis at Multiple Resolutions

· Support Modeling and Analysis across Security Domains Leverage Computation at Scale DoD’s digital strategy demands aggressive transformation of its underlying IT infrastructure and services in order to propel the enterprise into the age of model-driven decisions and artificial intelligence. The fractured landscape of legacy server racks and desktop enclaves have already begun coalescing into high-performance enterprise platforms comprised of scalable computing clusters, virtual desktop infrastructure, and commercial cloud services. Furthermore, advancements in compute, storage, and processing capabilities are transforming the manner in which the DoD engineers its warfighting solutions. DoD scientists and engineers have begun to explore massive tradespaces of possible design options using a “model-analyze-build" approach over the traditional “design-build-test" approach. AFSIM must anticipate and adapt to these paradigm shifts in order to remain relevant and useful to the S&T enterprise.

Under this initiative, the following focus areas prioritize investments that will improve the overall performance of the AFSIM framework so that it can leverage modern computational platforms and gracefully scale to address new workload demands characteristic of the exponential increase in multi-domain tradespace complexity.

· Improve Application Performance Monitoring

· Support Cloud-native Deployments

· Support Deep-Learning Applications of MS&A Deliver MS&A Capabilities at the Speed of Relevance Objective-3 reclaims agile development as a core competency of any modern S&T enterprise seeking to deliver paradigm-shifting tools and software-intensive capabilities at the speed of relevance.

Section 3.3.1 describes AFRL’s motivation and purpose for defining this objective, Section 3.3.2 describes its specific scope of focus, while Section 3.3.3 enumerates five key initiatives within that scope.

Purpose While most commercial sectors can quickly respond to dynamic market signals and deliver software products on a continuous cycle, the DoD has historically imposed antiquated lifecycle models on the research, development, procurement, and sustainment of software-intensive capabilities. These models assume that there are distinct points in the software lifecycle where investigatory activities begin and end, where development activities begin and end, and where traditional sustainment activities begin. By any modern standard, software is never done. The DoD’s ability to maintain operational superiority hinges upon its ability to continuously evolve its warfighting capabilities at the speed of relevance. Consequently, the S&T enterprise, as the DoD’s vanguard for technological dominance, must continuously evolve its engineering and analysis tools, like AFSIM, to accelerate the inception and delivery of multi-domain battlespace advantage.

Scope AFRL intends to improve its ability to quickly adapt to changing requirements and scale AFSIM development to support a community of users across the DoD. Under this effort, offerors will apply modern software development practices and tools in accordance with Agile and DevSecOps principles to establish the enduring capabilities described in Table 6.

Table 6: Objective-3 Scope

Promote Agility and Discipline at Scale AFRL desires to collaborate with offerors who genuinely share an Agile mindset. Agile values emphasize collaborative and continuous feedback and learning through frequent (i.e., iterative and incremental) delivery of working software. Offerors that smartly apply Agile principles and tailor their practices to suit the MS&A application domain, should fundamentally structure, plan, and execute their work differently. While Agile will not solve essential problems in software development, it can help make those problems more visible. Development processes should promote transparency and maximize chances for the team to “inspect and adapt” to evolving stakeholder needs, evolving technologies, and evolving programmatic opportunity. Furthermore, attention to quality should apply to all phases of development starting with the capture of user needs through frequent and incremental validation of potential solutions. Sharing many principles with “lean thinking”, Agile should demand careful attention to the flow of work and the aggressive elimination of wasteful activities that do not seek to maximize value through a minimum-viable product.

Under this initiative, the following focus areas prioritize investments that will establish Agile Coordination as an enduring capability necessary to deliver new AFSIM capabilities at the speed of relevance.

· Amplify and Integrate User Voices

· Tailor an Agile Approach for Scaled Development

· Adhere to Agile Planning and Requirements Management Build the Software Factory AFRL desires to judge a successful AFSIM enterprise by its capacity to continuously deliver working code and actual value to end-users within a relevant cycle time. Organizations conducting multi-domain analyses or transitioning solutions to the warfighter will not wait for AFSIM to address technical gaps and deficiencies. Rather, these organizations will pursue a path of least resistance, even if those decisions needlessly duplicate effort across programs. While AFSIM represents an enormous investment in potential cost-avoidance, organizations will not realize these benefits until AFRL establishes pipelines that can keep pace with a continuous delivery cadence. Furthermore, these pipelines should not sacrifice the ability to test and verify software against defined quality profiles for multi-level security environments. DevSecOps describes a culture and set of practices that help unify development, security, and support/operations activities while “left shifting” many related software lifecycle concerns. The AFSIM enterprise will succeed as a multi-domain enabler to the extent it can focus its discipline and agility through DevSecOps principles while migrating towards future “M&S-as-a-service" concepts.

Under this initiative, the following focus areas prioritize investments that will establish Agile Product Delivery as an enduring capability necessary to provide new AFSIM capabilities at the speed of relevance.

· Perform Continuous Incubation and Exploration

· Perform Continuous Inspection and Integration Scale Content Delivery and Operations While AFRL has distributed AFSIM to over 300 organizations and trained over 1500 AFSIM users and developers to-date, it has encountered many practical challenges when attempting to meet demand at DoD scale. Operational scale drivers include increased demand for training, increased frequency of end-user support requests, new requirements for specialized product variants (e.g., classified and FVEY software distributions), requests to deploy AFSIM on shared, multi-tenant computing infrastructure, and a growing backlog of requests to use AFSIM as a platform to distribute third party or IC-validated models and scenarios. While AFRL has started to make smart infrastructure and tool investments (Section 2.6) to modernize its enterprise, these investments require up-front planning and coordination and cannot succeed without full stakeholder commitment.

Under this initiative, the following focus areas prioritize investments that will establish Scalable Operations as an enduring capability necessary to deliver new AFSIM capabilities at the speed of relevance.

· Establish a Robust Content Delivery Network

· Diversify Documentation, Training, and Support Approaches Glossary Common Terms and Definitions