LA-UR-26-23318 Approved for public release; distribution is unlimited. Title: CONTINUOUS IMPROVEMENT PROCESS GUIDE Author(s): Santonastaso, Nathanael James Intended for: Report Issued: 2026-04-23 Los Alamos National Laboratory, an affirmative action/equal opportunity employer, is operated by Triad National Security, LLC for the National Nuclear Security Administration of U.S. Department of Energy under contract 89233218CNA000001. By approving this article, the publisher recognizes that the U.S. Government retains nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or to allow others to do so, for U.S. Government purposes. Los Alamos National Laboratory requests that the publisher identify this article as work performed under the auspices of the U.S. Department of Energy. Los Alamos National Laboratory strongly supports academic freedom and a researcher's right to publish; as an institution, however, the Laboratory does not endorse the viewpoint of a publication or guarantee its technical correctness. CONTINUOUS IMPROVEMENT PROCESS GUIDE by Nathanael Santonastaso Submitted to the Capstone Project Committee in partial fulfillment of requirements for the degree of Master of Engineering Technical Management Texas A&M University Professor: Dr. Chahriar Assad Sponsor: Thomas Zambo, First-line Manager, Los Alamos National Laboratory April 2026 Copyright © 2026. TABLE OF CONTENTS Page TABLE OF CONTENTS ................................................................................................................ 1 EXECUTIVE SUMMARY ............................................................................................................ 4 ACKNOWLEDGEMENTS ............................................................................................................ 6 1. INTRODUCTION .................................................................................................................... 7 1.1 Project Proposal ............................................................................................................ 7 1.2 Literature Review ......................................................................................................... 8 1.3 Business Case ............................................................................................................. 10 2. METHODS ............................................................................................................................. 12 2.1 Methodology ............................................................................................................... 12 2.2 Project Plan ................................................................................................................. 15 3. RESULTS ............................................................................................................................... 21 3.1 Data Analysis Report .................................................................................................. 21 3.2 Decision Making Report ............................................................................................. 26 3.3 Financial Analysis Report .......................................................................................... 31 4. DISCUSSION AND CONCLUSION .................................................................................... 34 4.1 Discussion ................................................................................................................... 34 4.2 Conclusion and Future Work ...................................................................................... 35 REFERENCES ............................................................................................................................. 37 APPENDIX A: PROJECT CHARTER ........................................................................................ 39 APPENDIX B: DATA ANALYSIS ............................................................................................. 43 LIST OF TABLES Table I: Interview Question Sheet .................................................................................... 13 Table II: Project Tasks ...................................................................................................... 16 1 Table III: 2x2 Risk Matrix ................................................................................................ 18 Table IV: Interview Question Sheet ................................................................................. 19 Table V: Key Manufacturing Indicators ........................................................................... 22 Table VI: Interview Data Summary .................................................................................. 23 Table VII: Improvement Ranking System ........................................................................ 24 Table VIII: Sample Data ................................................................................................... 24 Table IX: Sample Improvement Data ............................................................................... 25 Table X: Sample Data Projected Improvement ................................................................ 25 Table XI: Actual Project Improvement............................................................................. 25 Table XII: Interview Data Summary ................................................................................ 28 Table XIII: Actual Project Improvement .......................................................................... 30 Table XIV: Actual Project Improvement .......................................................................... 32 LIST OF FIGURES Fig. 1: Process Flow Labels [3] ........................................................................................ 14 Fig. 2: Schedule Gantt Chart ............................................................................................. 17 Fig. 3: Process Flow Labels [3] ........................................................................................ 20 Fig. 4: CI Process Flow Chart with number identifiers .................................................... 23 Fig. 5: Final CI Process Flowchart ................................................................................... 24 Fig. 6: Final CI Process Flowchart ................................................................................... 27 Fig. 7: Final CI Process Flowchart with Legs................................................................... 28 Fig. 8: REN-dependent steps in Process Flowchart .......................................................... 28 Fig. 9: CI Process Flowchart ............................................................................................. 32 2 Fig. A.1: Charter Page 1 ................................................................................................... 39 Fig. A.2: Charter Page 2 ................................................................................................... 40 Fig. A.3: Charter Page 3 ................................................................................................... 41 Fig. A.4: Charter Page 4 ................................................................................................... 42 Fig. B.1: Final Flow Chart ................................................................................................ 43 Fig. B.2: Sample Calculations .......................................................................................... 44 3 EXECUTIVE SUMMARY This project develops and pilots a Continuous Improvement (CI) Process Guide and supporting data‑capture system for Los Alamos National Laboratory (LANL) as it transitions from a research and development environment to a production environment. The core finding is that LANL’s existing approach to production improvements is informal and fragmented: improvement ideas move through different paths depending on which group is involved, ownership and approval requirements are often unclear, and critical knowledge is lost with employees changing roles because LANL is experiencing great growth. The project shows that by formalizing the CI process, defining clear roles, and using a simple, quantitative scoring system tied to key manufacturing indicators, LANL can prioritize higher‑impact improvements and combine improvement incorporation to reduce non‑conformances, process cycle time, paperwork, increase process success rate, and achieve a positive financial return. The literature indicates that organizations frequently struggle with CI because they chase easy wins, lack a structured method, and fail to build an organization‑wide CI mindset, which leads to short‑lived efforts and loss of momentum. Successful CI programs instead rely on broad participation, transparent data capture, clear responsibilities, and disciplined follow‑through on known improvements. These lessons guided the design of a mixed‑methods confirmatory study at LANL that combined qualitative interviews with quantitative performance data to create a practical, sustainable CI framework. The project’s methods focused on a single production inspection process as a pilot area. Qualitative data was collected through interviews with Design Agency (DA), Product Engineering (PE), Process Engineering (PrE), and Quality Engineering (QE) teams to map how improvements are currently proposed, reviewed, and implemented, and to gather a set of known improvement opportunities. This data was used to construct a detailed CI flowchart that traces more than 50 steps and handoffs across groups, and to develop a scoring system that rates each improvement on its expected impact on four key indicators. The scoring system was iteratively reviewed with representatives from each group to reduce bias and ensure that no single team’s priorities dominate the ranking. The results show that the new CI framework can convert scattered ideas into an ordered, defensible list of improvements. Twelve interviews produced 38 distinct improvement suggestions, focused on a specific inspection process, confirming both high need and broad support for change. Applying the scoring matrix identified one improvement as the highest priority. Analysis of sample and historical data indicates that implementing this improvement is projected to prevent 5 non‑conformances per year, reduce inspection cycle time by 37.3 percent, eliminate 166 hours of paperwork annually, and increase the success rate by 61.9 percent. The financial analysis translates these operational gains into a return on investment for LANL. The project required 172 hours of effort, resulting in a total fully burdened cost estimated 4 between 45,150 and 51,600 dollars, which is lower than the original 54,600 to 62,400 dollar cost estimate. Valuing only the 166 saved hours per year from the top‑ranked improvement, and using LANL’s burdened labor rate, yields annual savings of approximately 43,575 dollars and a three‑year cumulative savings of about 130,725 dollars. Comparing this benefit to the project cost results in a three‑year ROI of roughly 190 percent and a payback period of about 1.04 years, even before accounting for additional benefits from reduced non‑conformances and improved throughput. In conclusion, the project found that LANL would greatly benefit from adopting the Continuous Improvement Process Guide and the associated scoring and data‑capture tools as standard elements of its production environment. This framework provides a clear, shared method for recording and prioritizing improvements, reduces knowledge loss, and strengthens cross‑functional collaboration. The demonstrated ROI from a single high‑impact improvement, combined with the potential to apply the same approach to additional processes, indicates that continued use and expansion of this CI system will yield ongoing operational and financial benefits for LANL as it matures as a production process. 5 ACKNOWLEDGEMENTS Contributors I would like to thank my faculty advisor, Dr. Chahriar Assad, and my industry sponsors, Mr. Thomas Zambo and Mr. Ian McCarrel, for their guidance and support throughout the course of this research. Thanks also go to my friends and colleagues, and the department faculty and staff for making my time at Texas A&M University a great experience. Finally, thanks to my family for their encouragement and to my wife and son for their patience and love during this process. The interview data, continuous improvement suggestions, and feedback used for the Continuous Improvement Process Guide were provided by my fellow colleagues, including the Quality Engineering, Process Engineering, Product Engineering, Design Agency, and Inspection teams at Los Alamos National Laboratory. This information was vital to the project, and again, thank you all for the support and effort you committed to this project. All other work conducted for the thesis was completed by the student independently. Funding Sources This capstone project was completed as part of the Master of Engineering Technical Management program at Texas A&M University and did not receive any external research funding. Los Alamos National Laboratory did not provide additional financial support beyond allowing the use of my time and effort to complete this work. 6 CONTINUOUS IMPROVEMENT PROCESS GUIDE 1. INTRODUCTION 1.1 Project Proposal A. Introduction Los Alamos National Laboratory (LANL) is adapting from a research and design (R&D) facility to a production facility; with this transition, there has been an issue in implementing production improvements. The purpose of this project is to develop a continuous improvement guide that captures the necessary improvements and provides a framework for implementing them. B. Problem LANL is changing from an R&D facility to a production facility. This means that LANL’s mindset also needs to change. In an R&D mindset, innovation is prioritized over refinement and efficiency. Now that LANL is focusing on production, LANL needs a formal way to record improvements. Making changes to production requires multiple levels of testing and approval, depending on the process and its owner. The improvement needed for a given process may involve one or more stakeholders, each of whom may require a different approval process to implement the change. Identifying who owns the issues and who is responsible for implementing and approving changes to the production process is cumbersome and challenging, especially for new employees. Through normal attrition and adding numerous additional positions, resulting in numerous new hires. Without a formal system to capture improvement information, it is lost when a person leaves LANL or relocates to a new position within LANL, which means the new employee must rework the same issue again, rather than building on where the last person left off. C. Project Overview This project will create a space to record and track the progress of necessary improvements and develop a guide to identify the correct approval process for implementing these improvements. The first step will be creating a template in Excel to track known issues that need to be addressed. The continuous improvement (CI) coordinator will record suggestions and improvements, and employees within production will be able to make new suggestions. Each 7 major stakeholder group in production will need to be interviewed to gather initial improvements and identify the approval process for implementing them within each group. Once this information is collected, it will be sorted into a process flowchart. After the flowchart and data capture method are created, the CI coordinator will manage the list of improvements and ensure that the owning party actively addresses the recorded improvements and follows the correct approval process. D. Metrics The success of the project will be measured by tracking the current production Key Performance Indicators (KPIs) and comparing them to the KPIs after each improvement is made. The first phase of this project will involve determining manufacturing metric indicators that best serve the current lab mission and tracking and recording the current state of each, thereby creating a benchmark to review as new improvements are implemented. E. Summary This project will capture critical data to improve production by stopping unnecessary rework and creating a formal process to track and implement production improvements and to see their effects on KPIs. 1.2 Literature Review A. Background To build and implement a Continuous Improvement (CI) program at LANL, the motivation needs to be defined. Why is continuous improvement significant? Along with motivation, the various methodologies for continuous improvement need to be explored to determine which is best suited to be implemented at LANL to be an effective, sustainable, and successful. B. Why Companies Struggle to Implement Continuous Improvement? Implementing Continuous Improvement seems like an obvious thing for every company to do. Who wouldn’t want to get more for less, but in reality, implementing continuous improvement is difficult and requires a culture shift [1]. This means that there isn’t just a team working on CI, but the organization needs to have a CI mindset. Without a systematic approach, CI can lose support quickly [2]. Often, when CI is attempted, the easiest items are attacked first, which seems appealing; however, without a structured approach, the momentum from the early success will fade, leading to only short-term success. The lack of sustained results leads to a decrease in support for CI within the organization [2]. Without sustained support for CI, there is no time to build an organizational culture that understands the value and importance of CI. 8 C. Successful Continuous Improvement Methodologies The implementation of CI may vary between companies, but the methodology is similar across many organizations. CI is most successful when the entire company can participate in CI [3]. Successful CI programs take a structured approach [2]. A significant component of enabling LANL to move towards CI is creating a flowchart that guides each employee in participating in CI [3]. This flowchart will create a path to transfer knowledge between groups and act as an aid so that each team knows what the other team is responsible for and what is required for each team to implement an identified improvement. Defining roles and responsibilities and giving everyone access to the CI process is crucial for success [4]. Once the roles are well defined, a data capture will be created to record potential improvements. This allows a systematic approach to be taken on the implementation of the improvements. Rather than each group working blindly, there will be a single point where all information converges, and the entire organization can have access [1]. The final element that needs to be considered is the follow-through. The final step in CI is to follow through and implement the identified issues [5]. For CI to be successful long- term, there needs to be a plan for addressing the potential improvement that has been recorded. D. How to Implement Continuous Improvement Successfully. To successfully implement CI, everyone needs to be involved, and sufficient Benchmarking needs to be created. CI needs to be integral to the design of each process [6]. This project will act as a modified Kaizen event. A Kaizen event is a short-term, focused team with an accelerated timeframe to improve a targeted area of a company [7]. In this modified version, this project will act as a pilot focused on one area of production. Data will need to be collected throughout the project to measure success [8]. Creating the right benchmarking will be a pivotal element in the project’s success, so all available historical, current, and future data will need to be collected [9]. Group participation will be a key benchmark indicator for the project’s sustainability [10]. CI is only as good as the input it receives from the production floor. If certain groups aren’t participating, CI can hurt overall performance by handicapping a less active group, which will cause resistance to CI and ultimately lead to a failure to maintain a CI program. E. Expected Results The implementation of CI is associated with two primary benefits: increased production and decreased non-conformances [11, 12]. In one documented case, a manufacturing process experienced a 13.3% increase in cycle time, while specific process improvements resulted in efficiency gains of up to 50% [11]. Furthermore, establishing a sustainable CI program facilitates the identification of the root causes of non-conformances, allowing for the necessary adjustments to be made to reduce their occurrence [12]. 9 F. Implications The key implications that need to be highlighted are the importance of participation in the CI program. This project needs to be cross-functional between all the teams that support production. Each team needs the ability to make input and see how the CI will affect their process. This means that the data in SharePoint will need to be available to each team. This will create an environment where a CI culture can grow. Participation in the CI program will also serve as an indicator of the program's sustainability. The flowchart will act as a bridge between each team, enabling them to understand the process of implementing an improvement and the effort required to enact it. The final element that will significantly improve the project’s performance is leaning on a structured approach to CI priority. Just because something might be easy to implement does not justify it to be high on the priority list. When implementing each recorded CI, there needs to be a benefit analysis to determine which improvement will make the most of the available resources. Though initial results might be delayed, taking the time to formalize and structure a process for CI will enable the project to build a good reputation and build confidence within the company that the data being captured will result in beneficial action. 1.3 Business Case A. Problem LANL is transitioning from a research and development environment to a production environment, and this shift has exposed significant gaps in how production improvements are identified, documented, and implemented. In the current state, improvement ideas move through a convoluted and largely informal process, where ownership is often unclear, and the path to approval depends on which group happens to be involved. Multiple stakeholders, such as the Design Agency (DA), Product Engineering (PE), Process Engineering (PrE), and Quality Engineering (QE), may each require different levels of authorization, which makes it difficult for new and existing employees to determine who is responsible for approving a change. As LANL experiences immense growth and changes in staff members, critical knowledge is lost, forcing new employees to rediscover and rework problems that had already been identified instead of building on prior work. Without a formal, visible system for capturing and executing improvements, production struggles to refine its processes, protect knowledge, and consistently meet LANL’s evolving manufacturing mission. B. Deliverables This business case proposes two primary deliverables that together establish a structured CI system for production. The first deliverable is a Continuous Improvement Process Guide that clearly defines how improvement opportunities are recorded, evaluated, approved, and implemented, including the roles and responsibilities of each stakeholder group. The second deliverable is a data‑capture tool that serves as a central repository for improvement ideas, associated process data, and status tracking. The CI coordinator will use this tool to record 10 known issues, log new suggestions from the production floor, and ensure that each improvement is routed to the correct owner and follows the appropriate approval path. C. Proposed Execution The proposed execution follows a phased approach using a mixed‑methods methodology. In Phase 1, the team will work with leadership to identify key manufacturing indicators that best represent LANL’s current production priorities, such as paperwork and inspection cycle times, failed runs, and non‑conformance counts. Phase 2 will establish a baseline by gathering data on these indicators and creating the SharePoint environment, including the initial templates for recording process information and suggested improvements. In Phase 3, interviews with Design Agency, Product Engineering, Process Engineering, and Quality Engineering will be used to document the existing improvement‑incorporation process and to collect a first set of improvement opportunities, which will be organized into a process flowchart. Phase 4 focuses on developing and refining an improvement‑scoring system that aligns each suggestion with the key manufacturing indicators, using iterative feedback from the stakeholder groups to reduce bias and ensure that the scoring remains balanced. Finally, Phase 5 will use the guide and scoring system to implement selected improvements, monitor participation across groups, and compare post‑implementation performance against the established baseline. D. Improvement Scoring The improvement‑scoring approach is intended to move LANL away from an ad hoc method of choosing which ideas to pursue and toward a data‑driven prioritization framework. When an improvement suggestion is submitted to the data capture tool, it will be evaluated against the key manufacturing indicators, including its projected impact on cycle time, reduction of failed runs and non‑conformances, and potential effect on cross‑group coordination. Each improvement will receive a quantitative score that reflects both its expected benefit and its resource requirements, allowing the CI coordinator and leadership to rank improvements. This structure is designed to avoid the common pitfall of implementing the “easiest” changes first, regardless of their strategic value, and instead directs attention to the improvements most likely to advance LANL’s production mission. E. Return on Investment (ROI) The anticipated return on investment for the Continuous Improvement Process Guide is realized through increased production efficiency, reduced rework, and improved reliability in meeting program demands. By capturing and reusing improvement knowledge rather than allowing it to leave with individual employees, LANL can avoid repeated troubleshooting of the 11 same issues and shorten the time from problem identification to resolution. The standardized process and scoring system also support the “speed of trust” within the organization: as stakeholders see that recorded improvements are handled consistently and fairly, confidence in the CI program grows, which in turn encourages more participation and faster decision‑making. Over time, improvements in key manufacturing indicators, such as lower non‑conformance counts and shorter cycle times, translate directly into lower operating costs and greater throughput, yielding a positive financial impact that justifies the initial investment in developing and maintaining the CI infrastructure. F. Recommendation Based on the identified problem, the proposed deliverables, and the expected operational and financial benefits, the recommended course of action is to implement the Continuous Improvement Process Guide and associated data‑capture tool as a standard element of LANL’s production environment. Formalizing how improvements are recorded, scored, and approved will clarify issue ownership, strengthen cross‑functional communication, and significantly reduce the knowledge loss that currently limits sustained progress. This structured CI system aligns with the broader transition from an R&D mindset to a production mindset by emphasizing refinement, efficiency, and data‑driven decision‑making. Implementing the guide now will position LANL to improve production performance in a measurable way while building a durable culture of continuous improvement that supports long‑term mission success. 2. METHODS The purpose of this project is to implement a continuous improvement (CI) program that tracks production metrics to verify that production is improving with the incorporation of new improvements and to establish a data-driven system to prioritize their incorporation. To best support this, a mixed-methods confirmatory research design will be used to gather qualitative and quantitative data to build a process flow guide and a system to record and score known improvements, enabling better resource allocation to the most impactful issues. 2.1 Methodology A. Research Design The program director has emphasized a push to make data-driven decisions rather than the shotgun approach, which often results from working in an R&D environment. As LANL refines its production program, it will need to formalize its CI process. To initialize the project, both qualitative and quantitative data will need to be gathered. This will be done in a concurrent parallel design. The qualitative data will be the information gathered for the process flow, showing what each group goes through to incorporate a new improvement. This data will be 12 collected through interviews, and CI recommendations will also be collected for use later in the project. The initial quantitative data will be collected to create a benchmark for production to compare over the next year and used to create a scoring system for the new improvements. This means that this project will be a mixed-methods confirmatory study using both qualitative and quantitative data to verify the effectiveness of a formalized CI program. At this stage, a scoring system will need to be created that ranks each improvement previously collected against the key manufacturing data. This will be an iterative process that will require feedback from each group. This will be a key to avoiding bias from a specific group, especially the process engineer. This project focuses on one particular process within production, and to avoid Process Engineer bias, with the engineer as the process owner, feedback from the other groups will be crucial. After the scoring system is approved and all improvements are ranked, a summary of all potential enhancements will need to be prepared to calculate a projected ROI for the forthcoming year. B. Participants To gain process flow knowledge, the following groups will need to be interviewed, starting with the least available and proceeding to the most available. This means that the Design Agency would be interviewed first, then the Product Engineers, Process Engineers, and finally the Quality Engineers. Each of these groups will again be called on during the scoring system development to ensure no one team is favored over another. The project sponsor and the FLM for the Quality Engineering Group will have the final say on the scoring system in the event of a disagreement that requires a tiebreaker when weighing improvements. The Quality Engineering groups will have access to non-conformance data and will be relied on heavily to gather that data. The CI coordinator will be responsible for reviewing and collecting previous process data. C. Procedure A standard set of questions will be asked of each member of each group. Table I will be used during each interview to record responses and organize data from each group. Table I: Interview Question Sheet Participant Participant Questions Response Name Group What steps do you have to go through to implement a new improvement? What approvals are required for those improvements? How do you implement an improvement that is within another group’s ownership? What current improvements do you know of that you would like to implement? Why haven’t you implemented these improvements? 13 In parallel with the interview process, key manufacturing data will need to be collected for the past 12 months. Key manufacturing data will include the total number of processes causing non-conformances per part, process cycle time for each inspection, the number of failed runs, and the paperwork cycle time. This will include average non-conformance disposition time, process reevaluation time, and sales package completion time. The CI coordinator will work with the Quality Engineering group, which will have access to non-conformance data and paperwork cycle time data. The CI coordinator will have access to previous production data and will be able to review and collect process cycle time and failed inspection data. The CI coordinator will also be responsible for creating the initial draft of the CI scoring system and refining it based on feedback from each group. In the event of a group conflict, the Quality Engineering first-line manager will guide the group in breaking any ties. D. Data Analysis Plan The data analysis will be completed in three parts. The first two parts will involve gathering interview data and creating a process flow sheet that outlines the process for implementing and approving new improvements. The format of this process flow chart will use the four standard symbols. The four symbols are shown in Fig. 1. The flowchart will be created using Visio. Fig. 1: Process Flow Labels [3] 14 The second part of the data analysis will involve creating a benchmark using the previously gathered key indicators. The collected data will need to be formatted into the following metrics: average non-conformances per part over the last year, average cycle time per part over the previous year, average failed inspections due to process failures over the last year, and paperwork cycle time per part. A potential pitfall to be carefully avoided is that non- conformances resulting from a part defect must not count against the benchmark. So, organizing and filtering the data to include only results that pertain to the process will be crucial to ensuring the project has an accurate benchmark that serves as a foundation for the rest of the project. The next analytical step will be the creation of a scoring system. Each CI suggestion will need to be scored and rated on its impact on these key indicators. Once a rough draft of the scoring system is in place, it will be reviewed by a member of each group to identify potential bias and ensure no one type of issue is inappropriately over- or underweighted. Each CI suggestion will have an impact score to set its priority. Additionally, a summation of each improvement will be calculated so the potential of the process can be projected for a given improvement. This will be key for identifying paths forward and making process decisions to determine whether a current process has the potential to meet its performance targets, or whether a certain inspection needs to be reworked entirely. E. Summary This project will encompass both qualitative and quantitative data. It will confirm the hypothesis that production will improve by introducing a formal CI system that scores potential improvements and strategically incorporates them. This will require one-on-one interviews and looking at historical data over the last year to create a process benchmark and scoring system. This will allow production to strategically focus resources on the highest-impact issues within process improvement, improve overall production efficiency, and build greater trust within the National Laboratory Complex. 2.2 Project Plan A. Introduction The purpose of this project is to implement a continuous improvement (CI) program that tracks production metrics to verify that production is improving with the incorporation of new improvements and to establish a data-driven system to prioritize their incorporation. To best support this, a mixed-methods confirmatory research design will be used to gather qualitative and quantitative data to build a process flow guide and a system to record and score known improvements, enabling better resource allocation to the most impactful issues. The following sections will further outline the project plan and a detailed methodology for the project execution. 15 B. Scope The scope of this project is broken into eight major tasks. These tasks outline the project's scope and major milestones and are shown in Table I. Table II: Project Tasks Task # Description 1 Identify Key Manufacturing Indicators 2 Record QE, PrE, PE, & DA Processes 3 Create Data Capture 4 Gather Data on Key Indicators 5 Record Current Improvements 6 Create Process Flow Guide 7 Score Identified Improvements 8 Calculate Projected ROI The deliverables for the project will be a list of known improvements, ranked from most to least impactful, and a flow diagram illustrating the process for incorporating those improvements. The projected ROI will show the potential of the current process after the known deficiencies identified in this project are addressed. C. Schedule Fig. 1 is a Gantt chart outlining the project schedule. Any task number within a Gantt chart bar represents the task that must be completed before that task can start. 16 Identify Key Manufactureing Indicators (1) Record QE, PrE, PE, & DA Processes (2) Create Data Caputre (3) Gather Data on Key Indicators (4) Task 1 Recorded Current Improvements (5) Task 2 & 3 Create Process Flow Guide (6) Task 2 Score Identified Improvemeents (7) Task 5 Calculate Projected ROI (8) Task 7 Fig. 2: Schedule Gantt Chart D. Project Budget The initial project timeline will be from November 24th to January 29th. Excluding weekends and holidays, that is 480 working hours, and at 1/2 a full-time employee (FTE), that is 240 working hours. With a burden of 3.5 to 4 times the employee's salary, the estimated cost is between $54,600 and $62,400. Additional hours required for reviews and interviews will be recorded during the project to get a complete project cost. E. Resources To gain process flow knowledge, the following groups will need to be interviewed, starting with the least available and proceeding to the most available. This means that the Design Agency would be interviewed first, then the Product Engineers, Quality Engineers, and finally the Process Engineers. Each of these groups will again be called on during the development of the scoring system to ensure no team is favored over another. The project sponsor and the FLM for the Process Engineering Group will have the final say on the scoring system in the event of a disagreement that requires a tiebreaker when weighing improvements. The Quality Engineering groups will have access to non-conformance data and will be relied on heavily to gather that data. The CI coordinator will be responsible for project execution and coordinating with each group to gather the necessary data. The CI will also review and collect previous process data needed to create the process benchmark. 17 F. Risk Management Plan A significant risk factor is the inability to interview the required groups within the scheduled timeframe to obtain the information needed to create the flow diagram. To mitigate this risk, the groups with the least availability will be interviewed first, allowing the most flexibility to complete the necessary interviews. This will be the DA; the next least accessible group is PEs, then QEs, and finally the PrEs, which have the most access and will be prioritized last. Preliminary discussions to schedule interviews have already begun to mitigate this risk. The potential risk is that unforeseen bias may appear in the scoring system. When creating the scoring system, each production team will review the preliminary scores of each improvement to determine if the scoring system is appropriate. This may create tension and bias. The FLM and Sponsor are ultimately the tie-breakers; however, a highly experienced quality engineer, who does not belong to the process, may be brought in, with no bias towards a specific team, to provide additional clarity if needed. This engineer will act as an unbiased third party if need be. Among Scope, schedule, and cost, the highest-risk item that cannot be moved is the overall schedule. The project cost is minimal and does not require any additional direct funding. The CI coordinator can donate time to the project in the case of funding or project support issues throughout the execution phase. In the event of unforeseen risks, the CI coordinator will meet with the project sponsor to discuss a path forward and make necessary adjustments, including scope reduction. Preferably, each group would have multiple representatives to be interviewed to ensure the best possible representation of the process flow guide, and, again, for the iteration of the scoring system, feedback would be gathered from various representatives. However, if unforeseen issues create scheduling constraints, these interviews can be reduced to a single representative. During the scoring iteration process, a roundtable can be held with a single representative from each group to significantly reduce the time to finalize the scoring system and overall project time. Table III: 2x2 Risk Matrix Low Consequence High Consequence Cost / Funding Constraints: Bias in Scoring System: Unforeseen bias Minimal direct funding limits could distort priorities and damage trust. project flexibility, but the Mitigation: Have all production teams review Low impact is small. Mitigation: preliminary scores; use FLM/Sponsor as Probability Use the CI coordinator’s tie-breakers; bring in an external QE as an donated time to cover gaps unbiased third party. and adjust effort as needed. High Schedule Slip: With a single Inability to Complete Interviews on Probability CI coordinator, the schedule Schedule: Not all groups can be interviewed 18 Low Consequence High Consequence is likely to slip in high- in time, delaying the flow diagram and scoring priority events, preventing system. Mitigation: Interview the work on the project. least-available groups first; start scheduling Mitigation: Use the CI early; if needed, limit to one representative per coordinator’s donated time to group and use a roundtable session to finalize cover gaps and adjust effort scoring. as needed. G. Research Plan Methodology The program director has emphasized a push to make data-driven decisions rather than the shotgun approach, which often results from working in an R&D environment. As LANL refines its production program, it will need to formalize its CI process. To initialize the project, both qualitative and quantitative data will need to be gathered. This will be done in a concurrent parallel design. The qualitative data will be the information gathered for the process flow, showing what each group goes through to incorporate a new improvement. This data will be collected through interviews, and CI recommendations will also be collected for use later in the project. The initial quantitative data will be collected to create a benchmark for production to compare over the next year and used to create a scoring system for the new improvements. This means that this project will be a mixed-methods confirmatory study using both qualitative and quantitative data to verify the effectiveness of a formalized CI program. At this stage, a scoring system will need to be created that ranks each improvement previously collected against the key manufacturing data. This will be an iterative process that will require feedback from each group. This will be a key to avoiding bias from a specific group, especially the process engineer. This project focuses on one particular process within production, and to avoid Process Engineer bias, with the engineer as the process owner, feedback from the other groups will be crucial. After the scoring system is approved and all improvements are ranked, a summary of all potential enhancements will need to be prepared to calculate a projected ROI for the forthcoming year. Each available member of each group will ask a standard set of questions. Table II will be used during each interview to record responses and organize data from each group. Table IV: Interview Question Sheet Participant Participant Questions Response Name Group What steps do you have to go through to implement a new improvement? What approvals are required for those improvements? How do you implement an improvement that is within another group’s ownership? 19 What current improvements do you know of that you would like to implement? Why haven’t you implemented these improvements? In parallel with the interview process, key manufacturing data will need to be collected for the past 12 months. Key manufacturing data will include the total number of processes causing non-conformances per part, process cycle time for each inspection, the number of failed runs, and the paperwork cycle time. This will include average non-conformance disposition time, process reevaluation time, and sales package completion time. The CI coordinator will work with the Quality Engineering group, which will have access to non-conformance data and paperwork cycle time data. The CI coordinator will have access to previous production data and will be able to review and collect process cycle time and failed inspection data. The CI coordinator will also be responsible for creating the initial draft of the CI scoring system and refining it based on feedback from each group. In the event of a group conflict, the Process Engineering first-line manager will guide the group in breaking any ties. The data analysis will be completed in three parts. The first two parts will involve gathering interview data and creating a process flow sheet that outlines the process for implementing and approving new improvements. The format of this process flow chart will use the four standard symbols. The four symbols are shown in Fig. 3. The flowchart will be created using Visio. Fig. 3: Process Flow Labels [3] 20 The second part of the data analysis will involve creating a benchmark using the previously gathered key indicators. The collected data will need to be formatted into the following metrics: average non-conformances per part over the last year, average cycle time per part over the previous year, average failed inspections due to process failures over the last year, and paperwork cycle time per part. A potential pitfall to be carefully avoided is that non- conformances resulting from a part defect must not count against the benchmark. So, organizing and filtering the data to include only results that pertain to the process will be crucial to ensuring the project has an accurate benchmark that serves as a foundation for the rest of the project. The next analytical step will be to create the scoring system. Each CI suggestion will need to be scored and rated on its impact on these key indicators. Once a rough draft of the scoring system is in place, it will be reviewed by a member of each group to identify potential bias and ensure no one type of issue is inappropriately over- or underweighted. Each CI suggestion will have an impact score to set its priority. Additionally, a summation of each improvement will be calculated to estimate the potential of the process for that improvement. This will be key for identifying paths forward and making process decisions to determine whether a current process has the potential to meet its performance targets, or whether a certain inspection needs to be reworked entirely. H. Summary The project will be executed between November 24th and January 29th. In the event of unforeseen issues, the project scope will be reduced to maintain the project schedule. This project will encompass both qualitative and quantitative data. It will confirm the hypothesis that production will improve by introducing a formal CI system that scores potential improvements and strategically incorporates them. This will require one-on-one interviews and looking at historical data over the last year to create a process benchmark and scoring system. This will allow production to strategically focus resources on the highest-impact issues within the production process, improve overall production efficiency, and build greater trust within the National Laboratory Complex. 3. RESULTS 3.1 Data Analysis Report A. Executed Process The purpose of this section is to elaborate on the executed project plan and describe any changes that were made throughout the project execution, which included slight changes to the data collection. The first step in the project execution was to meet with the management team and identify the key manufacturing indicators most crucial to LANL’s manufacturing mission in its 21 current state, after meeting with the group leader (GL) and first-line manager (FLM), who had identified various key manufacturing indicators. Once the indicators were identified, interviews were conducted to gather process data on how to incorporate inspection improvements. The results of the interviews were a process flowchart of the improvement incorporation process. During the interview process, improvement suggestions were also collected for processing later in the project. One adjustment from the original planned execution was that the process information from the interviews was recorded directly into Visio. During the first interview, it became apparent that having the flowchart available to record data would be the most effective way to help the interviewee understand the flowchart's objective and obtain the intended feedback on process information. After each interview, the recorded suggestions were also scored using the scoring system, and the grading system was adjusted with each additional suggestion. This was done at the end of the interviews to condense the project schedule. Once the interviews were complete, all suggestions were scored with the latest revision of the scoring system, and the top-ranked suggestion was identified. The top suggestion was analyzed to observe the impact of incorporating it. Baseline data was gathered for the processes impacted, to be used as a comparison with those before and after the improvement was completed. Once the data were collected, the projected impact of incorporating the improvement was calculated. B. Collected Data The purpose of this section is to outline the raw data collected throughout the project, identify its origin, and prepare it for presentation and use later in the project. Table V shows the key indicators used. These were determined based on input from the FLM and the GL of the Quality Engineering group, in conjunction with the FLM of the Process Engineers. Table V: Key Manufacturing Indicators # Description 1 Paperwork Cycle time 2 Inspection Cycle time 3 Failed Runs 4 Number of Non-Conformances (NCR) The following data were collected through interviews with the Design Agency, Product Engineer team, Quality Engineer team, and the Processes Engineer team. Table VI below summarizes the interview data, showing which interviewee contributed to the flowchart and the suggestions collected. Fig. 4 shows the flowchart layout and numbers for each element, so the data can be referenced back to the interviewee who contributed it. 22 Table VI: Interview Data Summary Interviewee # Process Flow Suggestions Group(DA, PE, PrE, Contribution Contribution # QE) 1 2, 3, 7, 8, 13 3 DA 2 4-6, 9-24, 2,4 PE 3 4-6, 10-24, 32 PE 4 23-31,47 7 QE 5 32-46 PrE 6 16,17 PrE 7 19,20 PrE Operator 8 19,20 PrE Operator 9 21,22 QE 10 32-46 18,29-35 PrE 11 1, 5-15, 23-28, 38 PrE 12 36, 37 PrE CI coordinator 1, 49, 50, 51 PrE/QE Fig. 4: CI Process Flow Chart with number identifiers C. Data Analysis Results The purpose of this section is to discuss the analysis performed on the data collected in the previous section. After making minor edits to the flow diagram, including grammatical corrections and design edits, Fig. 5 shows the final flowchart. 23 Fig. 5: Final CI Process Flowchart See Appendix B.1, which includes a larger flowchart for better readability. The next stage of the data analysis was to identify the top rank improvement. Table VII shows the scoring method used to rank each improvement. Table VII: Improvement Ranking System Score 0 1 2 3 Impact No Impact Low Impact Medium High Impact Description Impact Number of No Impact 1 NCR 2-3 NCR 4+ NCR reduction NCRs reduction reduction Cycle Time No Impact 0 hrs. -0.5 hrs. 0.5 hrs. -1hrs. 1 hr.+ Reduction Paperwork No Impact 0-50 hrs. 50-100 hrs. 100+ hrs. Reduction Failed Runs No Impact 1-2 FRR 3-4 FRR 5+ FRR Reduced (FRR) Table VII was used to rank all 38 improvements collected during the interview stage. The ranking system was developed by the CI coordinator and underwent multiple iterations during data collection. Once the ranking was complete, the highest-impact improvement suggestion was identified. To properly analyze the impact of this improvement, 12 months' worth of data on the process most affected by the improvement was collected as a baseline for comparison. Table IV presents a collection of Sample data used to illustrate how the projected improvement was calculated. Table VIII: Sample Data Cycle time 10 hrs. Failed Runs 20 Runs Total Runs 100 Runs Success Rate 80% 24 NCRs 10 Paperwork Time Per NCR - PrE 1hr. Paperwork Time Per NCR - PE 6hrs. Paperwork Time Per NCR - QE 6hrs. Failed Run Paperwork PE 1hr. Failed Run Paperwork PrE 1hr. Failed Run Paperwork QE 1hr. Table IX shows the sample improvement. This data was selected at random to demonstrate what the actual top improvement might have looked like. This was done to protect manufacturing data. Table IX: Sample Improvement Data Cycle Time None FRR 10 Runs NCRs Prevented 5 Table X shows the projected impact of the sample data. The number of NCRs and Paperwork hours is calculated using simple addition. The improvement in success rate was calculated using equation 1. See Fig. B.1 for calculations of the projected impact based on sample data. Table X: Sample Data Projected Improvement Number of NCRs 5 Cycle Time Reduction Remains Constant Paperwork Reduction 80 hrs. Success Rate Increase 12.5% Old Value−New Value Percent Improvement = × 100% (1) Old Value Table XI shows the actual projected improvement calculated from performing the same analysis on the top-ranked improvement from the initial interviews. Table XI: Actual Project Improvement Number of NCRs 5 Cycle Time Reduction 37.3% 25 Paperwork Reduction 166 hrs. Success Rate Increase 61.9% D. Recommendations and Observations The results of this project are promising and will have a tremendous impact on the quality and consistency of the inspection team within the production line. Inspections within a production line are crucial to successful manufacturing, but they are often given a very short timeline within the production line because nothing is happening to progress production. The part is set aside to be inspected before moving forward. Progress for the sake of progress is the biggest advocacy of inspection. The potential impact of these improvements does not just make inspection look better on paper and move parts through faster, but it also builds confidence in the results that are produced. A greater confidence means rejecting bad parts becomes easier, and being able to determine the root cause of the issue elsewhere in the production line becomes more possible. Now that the process for incorporating improvements has been mapped out in the flowchart, the top-ranked improvements have been identified. With their projected impact calculated, the next step is to begin the incorporation process and walk through the flowchart. Once the suggestion is incorporated, a new baseline can be determined and compared with the projected data calculated here. This does two things: first, it helps determine whether the analysis performed here was effective, and second, it reinforces the idea of continuous improvement in a calculated way. 3.2 Decision Making Report A. Introduction: To further analyze the project data and how it will be applied to future actions, descriptive and prescriptive analyses will be used. Descriptive and prescriptive analyses provide a coherent decision-making framework that directly supports the goals of this continuous improvement project at LANL. First, characterizing the current inspection and improvement incorporation process and then using that information to prescribe specific actions, this strategy ensures that recommended changes are both data-driven and practical within the organization’s operational context and operational goals. B. Decision-making methodology This project uses descriptive and prescriptive analysis together to first map how inspection improvements are currently identified and implemented at LANL and then to select and justify which proposed changes should be implemented first. Descriptive analysis summarizes how the improvement process actually works today by mapping each step, role, and handoff in the CI flowchart and using interview notes to highlight where work holds up or creates unnecessary work. Using descriptive analysis, one can draw insights from the flowchart to better understand the current state and make decisions to best capitalize on it [1]. Prescriptive 26 analysis then builds on that foundation by applying a structured, criteria‑based decision method (such as scoring improvements against key performance measures and constraints) to answer “given what we know, what should we do next?,” enabling the project to prioritize improvement options, articulate expected benefits, and recommend concrete changes that are both data‑driven and operationally realistic[13]. The goal of this project is to outline how to improve the current process now and do so with a data-driven approach, which is why this decision-making method works so well for this project. Naturally, the improvement process should improve over time as new improvements are suggested and implemented. Still, the goal is to outline the current process and enable employees by giving them the process knowledge to incorporate improvements, and to do so with a sound rationale, by giving employees a method to rank each improvement idea they have. Descriptive and predictive analyses are essential to the project's success. C. Descriptive analysis Descriptive analytics is used to understand what has happened [14]. The project’s descriptive analysis established a clear picture of the “as is” system, aligning tightly with the project goal of improving inspection quality and efficiency. Interviews with DA, PE, QE, and PrE stakeholders were used to construct a detailed CI process flowchart. Shown in Fig.6. See B.1 for a more readable version of the flowchart. Fig. 6: Final CI Process Flowchart Observing the flowchart, five distinct legs can be identified (see Fig. 7), each largely owned by a separate group or organization. This structure explains why teams have deep knowledge of their own segment but limited visibility into adjacent steps, which in turn contributes to siloed decision‑making and slower cross‑functional coordination. 27 Fig. 7: Final CI Process Flowchart with Legs When analyzing the PrE sections of this diagram, there are multiple independent document review cycles. These documents are interrelated, and currently, the revision of one DOP, SPI, or SQA will trigger a review of the two. Looking at the impact of initiating a REN, if a REN can be avoided, there are 17 steps within that can be bypassed. Fig. 8 illustrates the steps that are activated when a REN is required. This shows how important it is to combine multiple improvements into a single REN, or if it is a simple change, minimize it to avoid a REN if possible, because of the large review burden that is created when required. Fig. 8: REN-dependent steps in Process Flowchart Reviewing Table XII. The majority of the improvement suggestions came from the process engineers. Looking into the suggestions specifically, there is a specific inspection process (Designated as Process A) that received the most improvement suggestions across all 12 interviews, showing that Process A needs attention. Table XII: Interview Data Summary Interviewee # Process Flow Suggestions Group (DA, PE, PrE, Contribution Contribution # QE) 1 2, 3, 7, 8, 13 3 DA 28 2 4-6, 9-24, 2,4 PE 3 4-6, 10-24, 32 PE 4 23-31,47 7 QE 5 32-46 PrE 6 16,17 PrE 7 19,20 PrE Operator 8 19,20 PrE Operator 9 21,22 QE 10 32-46 18,29-35 PrE 11 1, 5-15, 23-28, 38 PrE 12 36, 37 PrE CI coordinator 1, 49, 50, 51 PrE/QE With a total of 38 improvement suggestions focused on inspection processes, this shows strong support for improvements and a wealth of ideas for the team to make future improvements. This seems obvious, but with change, there is a risk that the changes make a situation worse. With a strong directive to increase capability and a strong pool of suggestions in such a short period of time, this supports the overall notion that an improvement process that captures and incorporates improvements is needed. Continuing with descriptive analysis of the project, but shifting from the raw data to observations during the data collection/interview process. There was strong support from each interviewee for the overall objective. It was clear within the first few interviews that each team is relatively siloed, meaning that each team had a great understanding of their own process, but the understanding of the adjacent teams was limited. Roles and responsibilities were also not obvious to interviewees when discussing who is responsible for certain actions. D. Prescriptive analysis Taking what was learned from the Descriptive analysis section, it is now time to focus on prescriptive analytics. Prescriptive analytics answers the what’s next questions associated with the information gathered so far [14]. Prescriptive analysis was used to determine which specific improvement should be prioritized and how it would likely affect the system. By developing a structured scoring matrix that evaluated the 38 improvement suggestions against their projected impact on NCR reduction, cycle time reduction, paperwork hours, and failed runs, with each criterion scored from 0 (no impact) to 3 (high impact). This effectively transformed descriptive metrics into a decision rule: suggestions that have a greater impact on key manufacturers' indicators received higher composite scores and rose to the top of the priority list. For the top- ranked improvement, the projected impact is in Table XIII. 29 Table XIII: Actual Project Improvement Number of NCRs 5 Cycle Time Reduction 37.3% Paperwork Reduction 166 hrs. Success Rate Increase 61.9% This data supported the implementation of these improvements. Because the highest‑scoring improvements were concentrated in Process A and many of them required a REN, the analysis indicated that bundling these changes into a single REN cycle would maximize impact while minimizing additional review burden. Further review of the flow chart shows that when an improvement requires an REN, this not only adds a potential of 17 steps in the flowchart, but these reviews require another laboratory review, and many of those steps will bring more logistical hurdles when trying to implement an improvement because they are physically located in another state in a different time zone. When analyzing an improvement suggestion that will require a REN, all other suggestions associated with that process should be incorporated to reduce additional work. The top-rated suggestion that was evaluated earlier will require a REN, so all suggestions associated with Process A should be looked at and rolled into this REN cycle, so that during this REN, as many improvements as possible can be implemented, making more efficient use of the DA and PA’s time. Similarly, analysis of the document‑revision paths showed that current dependencies between DOP, SPI, and SQA updates create unnecessary cascaded work when only one document truly needs revision. The current flowchart does not require DOPs, SPIs, and SQA documents to be updated in unison, but they are currently written in such a way that an SQA revision will always require a DOP revision. By updating the DOP and SQA to no longer being co-dependent, it will allow for quicker SQA and DOP turnaround and again make the improvement process more efficient, encouraging more improvements in the future. E. Summary The project shows that combining descriptive and prescriptive analysis offers a practical path for advancing continuous improvement within LANL’s inspection processes by grounding recommendations in both observed reality and quantified impact. Descriptive analysis first clarified the current state by mapping the CI flowchart into five distinct organizational legs, revealing how DOPs, SPIs, SQAs, and RENs interact within the flowchart and where redundant review cycles and handoffs slow progress. Interviews with DA, PE, QE, and PrE personnel produced 38 improvement suggestions, many focused on inspection Process A, which, along with observations of siloed expertise and unclear responsibilities, underscored both the desire and need for change and the work culture barriers that currently limit efficiency. Prescriptive analysis then transformed these insights into prioritized action by applying a structured scoring matrix to all 38 ideas against key manufacturing indicators (NCRs, cycle time, paperwork hours, and failed runs), elevating a top‑ranked improvement projected to cut cycle time by 37.3%, 30 reduce paperwork by 166 hours, and increase success rate by 61.9%. This evidence supports moving forward with that improvement, bundling all related Process A changes into a single REN to maximize the gain from the review burden required of a REN with the off‑site DA, and decoupling DOP, SPI, and SQA revisions where feasible to prevent unnecessary cascading updates. This descriptive‑to‑prescriptive decision-making supported a Lab-wide initiative to push for more data-driven decisions, and LANL can shift from being reactive to proactive through new improvements that will resolve issues now and make future changes easier. The impact of this will support a metrics‑based CI culture that systematically captures ideas, evaluates them against mission‑critical goals, and implements the highest‑value improvements first and the improvements associated with the same specific process associated with the highest value to improve efficiency. 3.3 Financial Analysis Report A. Introduction The continuous improvement project successfully designed and recorded a structured process flowchart and scoring system for inspection-related improvements at LANL. The purpose of this section is to discuss the financial aspect of the project and the associated benefits, whether intangible or tangible. Furthermore, this section will discuss the future projected benefits over the coming year. This will include a post-implementation analysis of the top‑ranked improvement, which shows large measurable gains, particularly in cycle time, paperwork hours, and success rate that can now be translated into financial returns. B. Project cost The original project plan estimated a total labor cost between $54,600 and $62,400, based on 240 hours of work. The implied hourly cost at the low and high ends is $227.5 per hour (1) and $260.00 per hour (2). 54,600 Hourly Costlow = = $227.50 per hour (1) 240 62,400 Hourly Costhigh = = $260.00 per hour (2) 240 In actual execution, scheduled project work was reduced from 240 hours to 140 hours, with 12 one-hour interviews and 10 hours of review, for a total of 172 hours (3). Total Hours = 140 + 12 + 10 = 172 hours (3) Using an average salary of $150,000 and a burden rate between 3.5 and 4 times salary, the fully burdened annual cost per full-time equivalent (FTE) is between $525,000 (4) and $600,000 (5). 31 Annual Cost3.5× = 150,000 × 3.5 = $525,000 per year (4) Annual Cost4× = 150,000 × 4.0 = $600,000 per year (5) Assuming 2,000 working hours per year, the corresponding hourly rates are $262.5 (6) and $300 (7) per hour. 525,000 Hourly Rate3.5× = = $262.50 per hour (6) 2,000 600,000 Hourly Rate4× = = $300.00 per hour (7) 2,000 Applying these hourly rates to the 172 hours actually committed gives the realized cost range. Actual Cost3.5× = 172 × 262.50 = $45,150 (8) Actual Cost4× = 172 × 300.00 = $51,600 (9) Thus, the actual cost of $45,150 (8) –$51,600 (9) is below or at the low end of the original $54,600 to $62,400 estimate, indicating the project was completed under the initial cost projection. C. Tangible benefits The purpose of this section is to discuss the tangible benefits consisting of direct benefits and long-term projected benefits over the next year. As a direct result of this project's completion, the takeaways are the process flowchart Fig. 9 and a list of 38 recorded improvement suggestions, ranked by their impact on the inspection process. Fig. 9: CI Process Flowchart The top-ranked improvements were also identified, and the potential impacts of incorporating them were calculated. Table 1 shows the calculated potential gains. Table XIV: Actual Project Improvement 32 Number of NCRs 5 Cycle Time Reduction 37.3% Paperwork Reduction 166 hrs. Success Rate Increase 61.9% Now, to assess the project's future impact, we can calculate the expected 3-year ROI from incorporating the improvements, based solely on time savings. Cycle time and success rate will also have a financial impact, but due to laboratory restrictions, ROI will be based solely on time savings. The annual savings from 166 hours at a burden rate of 3.5 times the average salary of $150,000 are $43,575 (10), and the project cost was $45,150. Annual Savings3.5× = 166 × 262.50 = $43,575 per year (10) The three-year cumulative savings are $130,725 (11). Total Savings3yr, 3.5× = 3 × 43,575 = $130,725 (11) Total Savings3yr − Project Cost ROI3yr = (12) Project Cost Three-year ROI at a burden rate of 3.5 times the average salary is 190% (13). 130,725 − 45,150 85,575 ROI3yr, 3.5× = = ≈ 1.90 ≈ 190% (13) 45,150 45,150 Payback period (in years) is 1.04 years. Project Cost Payback Period = (14) Annual Savings 45,150 Payback3.5× = ≈ 1.04 years (15) 43,575 For 4 times the burden rate, the payback period and the ROI are the same as with the 3.5 burden estimate. These estimates also assume that over the next three years, production numbers will not increase, which we know is false. Though exact numbers are unknown, this would be the minimum expected ROI, indicating a strong return for the time invested in making these improvements. D. Intangible benefits Intangible benefits include a more mature CI culture, clearer ownership, and better cross‑group understanding of the inspection and improvement process, which was previously fragmented and siloed. The project lays the foundation for a shared CI mindset by giving DA, PE, QE, and PrE teams a common flowchart and language for how inspection improvements 33 move from idea to implementation. Having a formal system set up also starts to build confidence within the production line. The Inspection team takes a lot of scrutiny, especially when there are known issues within the inspection process. Having these issues documented, with a path forward identified, takes some of the pressure off when dealing with issues that take time to resolve, boosting confidence within the inspection team. Looking forward, a sustained CI culture will continue to strengthen the inspection process by making it more robust, more efficient, and less dependent on individual memory or ad hoc decision‑making. Over the long term, consistently meeting or exceeding internal performance objectives builds confidence not only within LANL but across the broader National Laboratory complex, giving LANL a reputational advantage when competing for new DOE contracts and demonstrating that it manages production quality through disciplined, data‑driven improvement rather than reactive fixes. E. Conclusion The CI project demonstrates that a relatively modest investment can generate substantial and sustainable value for LANL. By completing 172 hours of work at a fully burdened cost between $45,150 and $51,600, below the original $54,600 to $62,400 estimate, the project delivered a formal inspection‑improvement flowchart, a ranked list of 38 improvement opportunities, and a validated scoring method that directly links improvement ideas to measurable performance gains. The top‑ranked improvement alone is projected to prevent 5 NCRs, cut cycle time by 37.3%, reduce paperwork by 166 hours per year, and increase success rate by 61.9%, producing enough annual labor savings to achieve payback in just over one year and a three‑year ROI of roughly 190% based on time savings alone. From an equally crucial perspective, the project has begun to shift LANL from reactive, ad hoc fixes toward a disciplined CI culture with clearer ownership and a shared understanding of processes across DA, PE, QE, and PrE groups. The new CI flowchart and decision framework create not only a more robust and efficient inspection process today, but they also create a repeatable way to identify, prioritize, and justify future improvements as production demands grow. As these practices mature and additional high‑impact improvements are implemented, LANL can expect both tangible financial returns and the intangible cultural benefits to compound, which will strengthen the inspection performance within the production line, build confidence within the group and across the National Laboratory complex, and improve LANL’s competitive position when pursuing future DOE work. 4. DISCUSSION AND CONCLUSION 4.1 Discussion The literature on continuous improvement emphasizes that many organizations fail to sustain CI because they rely on ad hoc problem solving, chase easy wins, and do not create a common, structured method that withstand excessive company growth. Currently, improvements 34 in the inspection process have historically been handled through informal networks, with each group following its own path for proposing and implementing changes, which leads to unclear ownership, inconsistent follow‑through, and repeated work when people move roles. In contrast, successful CI programs stress broad participation, clear roles and responsibilities, and transparent data capture tied to measurable performance indicators. The CI flowchart developed in this project directly responds to those findings by creating a single, shared flowchart for implementing improvements and a data‑driven scoring system that links each suggestion to key manufacturing metrics, such as non‑conformances, cycle time, paperwork effort, and failed runs. The results confirm that this structured approach can convert scattered, qualitative feedback into a prioritized portfolio of improvement opportunities that is defensible to management and understandable to operators. Twelve interviews across DA, PE, PrE, and QE generated 38 improvement suggestions, many concentrated on a single inspection process, indicating a willingness to engage if a clear path is provided. The iterative development of the flowchart with each group, and later with subject‑matter expert feedback, not only produced a more accurate representation of the real process but also helped build early buy‑in for using the flowchart as the standard reference for future improvements. Likewise, the scoring system started as a simple matrix but evolved through cross‑team review into a tool that balances different perspectives and reduces the risk that a single group can dominate priorities based on local concerns rather than overall mission impact. For the financials and operations, the most important implication is that a relatively modest investment in formalizing the CI process can unlock substantial and recurring benefits. The top‑ranked improvement, focused on inspection, is projected to reduce inspection cycle time by 37.3 percent, prevent 5 NCRs per year, eliminate 166 hours of paperwork annually, and increase success rate by 61.9 percent. When even just the time savings are monetized using a fully burdened labor rate, the three‑year ROI is approximately 190 percent with a payback period of about 1.04 years, which provides a strong quantitative justification for moving this improvement, and the CI framework itself, into regular operations. In addition, the flowchart have already improved cross‑group understanding of who owns each step and what approvals are required, which supports the broader cultural shift from an R&D mindset to a production mindset. 4.2 Conclusion and Future Work This project effectively tested whether a formal CI framework, consisting of a shared process flowchart, a centralized data‑capture tool, and a quantitative scoring system, could be used at LANL to prioritize and justify production improvements, starting with inspection. The findings indicate that the framework is both technically sound and operationally acceptable: it produces clear, measurable projections of improvement impact, aligns with existing key performance indicators, and has been validated with stakeholders from DA, PE, PrE, and QE. As such, the project provides a strong basis for transitioning the CI Process Guide from a capstone deliverable into a standard operational tool, first within inspection and then, if successful, across additional production processes. 35 Several new project opportunities emerged from this work. First, the current flowchart and scoring system are focused on one inspection process; future projects could expand the same methodology to other operations. Second, once more improvements are implemented, there is an opportunity to conduct a longitudinal study comparing projected gains to actual gains, which would refine the scoring weights and further increase confidence in the tool for decision‑making. Third, the data‑capture and tracking system could be enhanced with simple dashboards that automatically display CI pipeline status, realized benefits, and group participation rates, turning CI into a visible management system rather than a background activity. At a broader level, this project reinforces lessons that are applicable beyond LANL. Organizations transitioning from R&D to production, or simply trying to scale operations with limited resources, benefit from treating CI as a structured, cross‑functional process rather than a side activity handled informally. It is possible to use metrics to build and test a scoring system, making the method adaptable to other highly regulated environments. Going forward, the plan is to continue to work with management to implement the top‑ranked improvements using the flowchart, measure actual performance changes against the projected baseline, and then use those results to advocate for expanding this CI framework across additional processes and, ultimately, across the broader LANL complex. 36 REFERENCES [1] S. Birckhead, "THREE KEYS TO CREATING A CONTINUOUS IMPROVEMENT CULTURE," Quality Progress, vol. 48, no. 7, pp. 32–37, Jul. 2015. [2] J. Ahlström, “Book Highlight-Sustaining the Continuous Improvement Culture and the Five Cultural Pitfalls Preventing You from Succeeding,” Global Business & Organizational Excellence, vol. 34, no. 5, pp. 62–72, 2015. [3] S. E. Burke and R. T. Silvestrini, Eds., The Certified Quality Engineer Handbook, 4th ed. Milwaukee, WI, USA: ASQ Quality Press, 2017. [4] S. Mizuno, Management for Quality Improvement: The 7 New QC Tools. New York, NY, USA: Productivity Press, 1988. [5] B. J. Schroer, M. Adams, S. Stewart, and P. J. Componation, “Continuous process improvement the quick step way,” Quality Progress, vol. 31, no. 2, pp. 85, Feb. 1998. [6] K. Ishikawa, What Is Total Quality Control? The Japanese Way, D. J. Lu, Transl. Englewood Cliffs, NJ, USA: Prentice-Hall, 1985. [7] W. J. Glover, J. A. Farris, and E. M. Van Aken, "The relationship between continuous improvement and rapid improvement sustainability," International Journal of Production Research, vol. 53, no. 13, pp. 4068-4086, 2015, doi: 10.1080/00207543.2014.991841. [8] V. M. Gilbuela-Subala, "Trails, Barriers, and Proposed Solutions in the Implementation of Continuous Improvement Project," Psychology & Education: A Multidisciplinary Journal, vol. 42, no. 2, pp. 196–255, 2025. [9] F. Roth, G. H. Ruff, and M. Sweeney, "Benchmarking Case Studies: Journeys Into Continuous Improvement," Journal: American Water Works Association, vol. 112, no. 1, pp. 75–77, Jan. 2020, doi: 10.1002/awwa 1435. [10] D. Jurburg, E. Viles, M. Tanco, R. Mateo, and Á. Lleó, "Measure to succeed: How to improve employee participation in continuous improvement," Journal of Industrial Engineering and Management, vol. 9, no. 5, pp. 1059–1077, 2016, doi: 10.3926/jiem 2074. 37 [11] A. Agustina, A. Widyarto, and T. Kriswanto, "Implementation of Continuous Improvement to Increase the Efficiency of the Production Process through Press Dies Design," International Journal of Research and Innovation in Social Science, vol. IX, no. IX, pp. 526–533, Sep. 2025. doi: https://dx.doi.org/10.47772/IJRISS.2024.809056 [12] B. Parente and A. Rocha, “Continuous improvement project to enhance product manufacturing quality in an industrial machine,” Procedia Computer Science, vol. 256, 2025, pp. 1772–177, 2025 [13] University of Bath, “Descriptive, Predictive, and Prescriptive Analytics,” University of Bath Online, 2025. [14] C. Cote, “Types of Data Analysis,” Harvard Business School Online – Business Insights Blog, 2019. [Online]. Available: https://online.hbs.edu/blog/post/types-of-data-analysis. Accessed: Mar. 4, 2026. 38 APPENDIX A: PROJECT CHARTER Fig. A.1: Charter Page 1 39 Fig. A.2: Charter Page 2 40 Fig. A.3: Charter Page 3 41 Fig. A.4: Charter Page 4 42 APPENDIX B: DATA ANALYSIS Fig. B.1: Final Flow Chart 43 Fig. B.2: Sample Calculations 44