DAI RPA Support JETS EFO Final Comp Ad Signed Redacted.pdf

Source Selection Information Page 2 See FAR 2.101 & 3.104

3. A DESCRIPTION OF THE SUPPLIES OR SERVICES REQUIRED TO MEET THE

AGENCY’S NEEDS/ESTIMATED DOLLAR AMOUNT:

DLA has a requirement for consulting services to strategically roll out the UiPath tool as well as support services to set up the implementation and deployment framework (i.e. screening criteria for processes appropriate for RPA, RPA design guidelines, business rules for deployments of RPA). These services will automate processes across the DAI Program Management Office (PMO) and agencies who are currently or planning on using RPA. Implementing RPA will achieve resource efficiencies in DAI’s customer-facing Sustainment Activities and some internal processes for development, test, and deployment of Problem Reports (break-fixes) and Patches (break-fixes from Oracle for Commercial Off the Shelf (COTs) issues.

A DAI Program task order (SP4709-19-F-0075) was awarded to International Business Machines Corporation (IBM) on September 2019 for RPA design, deployment, training, and sustainment activities. The requirement included the identification, development, and deployment of attended RPA bots that require human interaction to start the RPA script. At the time, the Government expected that 15-20 RPA candidates to be identified, developed, and deployed during the first year with a focus on training and sustainment task in the option years.

Since the award of task order SP4709-19-F-0075, the Contractor and PMO have identified over 60 additional RPA and the list of potential candidates continues to increase. Additionally, the Department focus is on further developing unattended RPA. Currently task order SP4709-19-F- 0075 is in option period one, which ends on September 29, 2021. This new task order will provide three full years of development activity, RPA identification, and a limited pilot of approximately five (5) unattended RPA bots candidates that run automation that works on its own, while maintaining the training and sustainment tasks. No disruption in service is anticipated.

Currently, the Government DAI Program Management Officer. provides contractors with two Government Furnished Equipment (GFE) tools: 1) Serena Business Manager which is used to receive, prioritize and track Production Service Requests from users, and 2) Serena Dimensions, which is used to maintain Configuration Control and Status reporting for ‘ready to deploy’ changes to the production application and status Accounting of the production environment.

Both tools require high levels of manual input and manual interface steps performed by the contractor. These tools currently do not support the entire Sustainment process flow from identified problem through root cause analysis, design, development, test, and deployment of break-fixes. The required RPA tool will allow the contractor to perform that function to script many of the manual process inputs to both of the GFI tools (Serena Business Manager and Serena Dimensions) as well as the manual steps external to the Serena tools which are part of the normal workflow for releasing code and configuration changes into the production.

A three year period of performance is required because it is tied to work that is being supported on the DAI Procure to Pay (P2P) contract (SP4701-18-C-0048) which has 3 years remaining (if options are exercised) in its period of performance. Year one and two will support the basic framework, documentation and implementation set up of RPA with attended bots. This includes identification, design, development, and deployment. Year three will include conducting an

Source Selection Information Page 4 See FAR 2.101 & 3.104

4. IDENTIFICATION OF THE EXCEPTION TO FAIR OPPORTUNITY (SEE FAR

16.505(b)(2)) AND THE SUPPORTING RATIONALE, INCLUDING A

DEMONSTRATION THAT THE PROPOSED CONTRACTOR’S UNIQUE

QUALIFICATIONS OR THE NATURE OF THE ACQUISITION REQUIRES USE OF

THE EXCEPTION CITED:

In accordance with FAR 16.505(b)(2)(i)(B), an exception to the fair opportunity process is justified because only one awardee is capable of providing the services required at the level of quality required because the services are unique and highly specialized. The DAI PMO required services are unique and highly specialized, as the vendor must have expertise and experience of deploying the Global Model for the DAI program as well as experience and expertise in designing, building, and managing RPA for the federal government. IBM has the unique experience implementing RPA which has been demonstrated with the work they performed with the Department of Homeland Security and the implementation/automation of over 50 processes.

None of the other current vendors who support the DAI program have implemented RPA. IBM is the only known vendor who has provided the required services to the federal government and exhibits expertise in the unique access control, system security and federal financial accounting laws, policies, and regulations.

The RPA services are required for a period of three years because the services are tied to work that is being supported on the DAI P2P contract (SP4701-18-C-0048) which has three years remaining (if options are exercised) in its period of performance. The underlying application design and configuration work on the P2P contract will define or constrain the actual process for which RPA will be appropriate and details of those actual designs are necessary inputs to the design and scripting of each RPA. Design and performance risk of separating these two tasks is high and even if successful would have inherent inefficiencies in required knowledge and technical data transfer that would likely offset the value of pursuing automation. Future re-competition of the P2P contract will include RPA within that scope.

The required support also includes providing expert support on the Oracle EBS Federal Financials DAI Global Model Platform (v12.2.6) combined with RPA to improve overall PMO operations, increase bandwidth of existing DAI resources, and reduce costs/re-work to deploy capabilities to the Other Defense Agencies utilizing DAI as their financial management Enterprise Resource Planning (ERP) at scale. The DAI PMO requires a vendor with highly specialized knowledge of DAI and the ability to develop automations using UiPath. The UiPath software has already been procured and is used by OSD/C and DAI, which allows for automation across the program. The vendor would be required to deploy and manage attended and unattended bots at scale, establish an Automation Center of Excellence (COE) and keep pace with the office of the Secretary of Defense Comptroller (OSD/C) and their efforts to automate financial management processes executed by the DoD Defense Agencies, under the Global Model of DAI.

Additionally, the vendor must have an in-depth understanding of the DAI operating model and the differences between the 27 agencies currently using DAI. The vendor will be required to establish a governance framework that will oversee enterprise automations and reduce the chance of duplicative efforts across DAI customers.

Source Selection Information Page 5 See FAR 2.101 & 3.104

IBM is the only known vendor that has an established process model that allows DAI to monitor and measure the performance of the UiPath software. In addition, while multiple vendors have DAI experience, IBM has a unique and highly specialized knowledge and capability with regard to the DAI applications development and support requirements based on their extensive knowledge and experience with the program application and operating environment, which is essential to implementing the RPA tool effectively. No other vendors with DAI experience possess the knowledge to effectively implement RPA. IBM is the only vendor who possesses the ability to provide DAI with strategic guidance necessary to establish governance, framework, and highly detailed processes in the automation processes across all the 27 DAI agencies. IBM has proven this unique skillset with the implementation of over 50 RPAs for the Department of Homeland Security. IBM is the only known vendor with DAI experience that has implemented

RPA.

IBM is the only vendor currently in a position to provide services at the necessary level of quality required for mission success. IBM is the only vendor that has the combined expertise needed to satisfy these requirements. IBM was the original vendor responsible for design and implementation of the DAI global model which has been a large factor in the success of the DAI program to date. IBM possesses a strong understanding of the functional requirements of the DAI PMO, its operations, the DAI technical architecture, and the priorities of its customer base, which is essential to successfully implement RPA into the DAI environment. The combined knowledge with RPA and the DAI Global Model is unique to IBM.

In addition, IBM has demonstrated experience in deploying UiPath RPA in the Federal government along with designing and implementing the governance and management structure for RPA programs in the Federal Government. Specifically, IBM has supported DAI since the beginning of the DAI program and has proven RPA experience dating back three years with implementing RPA for the Department of Homeland Security. To enable another vendor on RPA would require extensive training. However, even with training, another vendor would not have real world experience in deploying RPA in a production environment which would present significant risk and increased costs to the government. In addition, the PMO does not have the knowledge internally to be able to train a new vendor on establishing governance and management structures to support RPA. Since this is the PMOs initial implementation of RPA, DAI does not have the resources to train another vendor on RPA.

Lastly, training a vendor with RPA experience to understand the technical and functional details related to the global model would take at least six months to a year and require government resources to utilize time training the vendor in lieu of performing other mission critical tasks.

IBM’s unique expertise with DAI and RPA will allow it to provide the Government the following:

Optimize development operations of Oracle Financials ERP and related reporting environments

Optimize help desk activities for DAI by reducing response and resolution times, and significantly reducing human capital to resolve help desk tickets.

Sustainably scale for use against the DAI system, including management of identities, controls, and visibility

Source Selection Information Page 6 See FAR 2.101 & 3.104

Establish an Automation COE and governance model for the DAI Program to avoid unnecessary and duplicative automation.

No other vendor possesses or has demonstrated the required knowledge of both DAI and scalable robotic process automation specific to the DAI program. This includes the ability to apply screening criteria to potential RPA candidates (which requires significant technical design knowledge of DAI) and the ability to model potential automations to identify efficiencies (which requires both expert knowledge of the RPA tool and the applicable accounting law, policy and regulations). In addition, this includes the expertise in designing the appropriate governance rules and technical architecture for the reusability of assets and objects, as well as capabilities to monitor performance for advanced error and exception handling. No other vendor can offer the required services at the required level of quality because the services are unique and highly specialized.

Based on the facts above, and to ensure continuity of the required DAI global model support, this exception to the fair opportunity process is justified because IBM is uniquely qualified to meet the Governments need’s and ensure successful deployment of the UiPath RPA tool.

5. A DETERMINATION BY THE CONTRACTING OFFICER THAT THE

ANTICIPATED COST TO THE GOVERNMENT WILL BE FAIR AND REASONABLE:

As the Contracting Officer, I hereby determine that the anticipated cost to the Government will be fair and reasonable. The proposals will be analyzed in accordance with FAR 15.404-1 procedures. It is anticipated that the Government will use various price analysis techniques including, but not limited to, (1) comparison of proposed prices with the Independent Government Cost Estimate, (2) comparison with competitive price lists; and (3) comparison of proposed prices obtained through market research for the same or similar items in order to obtain a fair and reasonable price.

6. ANY OTHER FACTS SUPPORTING THE JUSTIFICATION:

OSD/C has undertaken an initiative to use RPA for the purpose of improving financial management practices in the Department of Defense. OSD/C is planning to use RPA to interact with DAI. It is paramount that OSD/C and DAI PMO RPA activates are done in a way that protects the performance and security posture of the Oracle EBS Application and DAI applied customizations. In addition, it is the goal of the DAI PMO to align with OSD/C’s initiative and to have an accelerated design/develop/deploy methodology for any RPA intended for deployment in the Oracle EBS application. It is imperative for RPA governance, design guidance, and RPA design/deployment to synchronize with this accelerated cycle. IBM’s combined expertise with DAI and RPA will allow the Government to do so in an accelerated fashion compared to any other vendor that would require extensive training, ensuring the OSD/C activities do not degrade the DAI system environment. Creating RPAs for DAI requires extensive knowledge of the Commercial Off the Shelf (COTS) application (Oracle EBS) and the ~170 DAI PMO developed custom design objects. Failure to fully understand or properly interpret system design documentation (delivered under the P2P contract SP4701-18-C-0048) will result in inefficient RPAs, which will degrade system performance when deployed.

Source Selection Information Page 7 See FAR 2.101 & 3.104

In addition, IBM possesses a pre-built virtual agent asset built specifically for the Government using Watson Artificial Intelligence (AI), known as Cognitive Automated Response Learning Agent (CARLA). CARLA is an IBM services asset that will be used to accelerate RPA engagements. CARLA serves as a digital mechanism to structure data from free form conversation. This is helpful when dealing with sustainment type activities and cuts down on the time required to get data in a format that RPA can use to execute an automated transaction. This combined AI asset for RPA along with IBM’s expertise deploying the DAI global model is what makes IBM’s support unique and highly specialized and what no other vendor possesses. IBM’s combined expertise of the DAI Global Model and RPA make its AI asset unique to DAI’s requirement. The AI is trained to understand customers’ requests and intent, and investigate issues, such as password issues, system errors, or access denials. The asset interacts with users or users’ systems to gather information necessary to trigger an unattended RPA bot to execute and close issues.

Award to a source other than IBM would result in a significant duplication of costs for another source to build a "CARLA-like" asset from scratch to work as required in the DAI environment.

In addition, IBM has developed RPA proofs of concept for assigning roles and responsibilities to users, as well as how to open and close periods (Acquire to Retire (A2R), Projects (PA), General Ledger (GL), and Purchasing) for Test and Development instances while supporting the DAI program, which another vendor does not possess. These are hurdles another vendor would need to overcome to successfully perform the requirement and would include learning the DAI global model, learning the DAI configuration and management processes, and/or learning how to design, build and manage RPA using UiPath in the DAI environment. This would create unacceptable delays (minimum of 6 months) in fulfilling DAI's requirements, while attempting to keep pace with the overarching financial enterprise RPA efforts led by the OSD/C.

OSD/C has created an initiative to increase enterprise automation for the Fourth Estate financial community, which will be led by DAI. This automation is required immediately, and IBM brings forth an immediate level of knowledge and expertise necessary from day one that can manage engagements with at least 30 unattended concurrent automations. The immediate need is because OSD/C has an already established RPA program and is currently creating RPAs that are available to use. It is essential that the DAI PMO quickly establishes an RPA framework to ensure no degradation to system performance or duplication of effort, as well as reviews all OSD/C RPAs to ensure the DAI Global Model is considered. Without this immediate support, the DAI program is at risk of increased system helpdesk tickets and performance issues.

7. A STATEMENT OF THE ACTIONS, IF ANY, THE AGENCY MAY TAKE TO

REMOVE OR OVERCOME ANY BARRIERS THAT LED TO THE EXCEPTION TO

FAIR OPPORUNITY BEFORE ANY SUBSEQUENT ACQUISITION FOR THE

SUPPLIES OR SERVICES IS MADE:

DLA has and continues to exercise competition as often as possible and is required to conduct ongoing reviews to keep abreast of new technology, both in industry and in the Government. As a continuing effort, DLA uses a variety of vendors, manufacturers, and contracts to fulfill its requirements and will continue its practice of surveying the market to ensure full and open competition. This task order will add RPA features in support of the DAI contract (SP4701-18-