A Lean Procurement Project at Chicago Public Schools

Assess the impact of constraints then follow a structured method to design a process

Published: Wednesday, August 21, 2013 - 15:36

Chicago Public Schools (CPS), the third largest school district in the United States, decided to review the way it procures items and services throughout a school year with the goal of saving money. At the beginning of 2012, the district’s chief operating officer announced a “lean review” initiative with the goal to uncover wasteful practices and unnecessary complexity, and to suggest actions to rectify them.

There were two main outcomes desired from the project. First was to make doing business with the district’s central offices faster and less expensive for CPS and its suppliers by examining how the procurement process could be accelerated. Second was to enhance the procurement process to capture information to be used by the district when making budget decisions to reduce expenditures by $100 million during the next three fiscal years.

Actions

Lean Six Sigma methods were employed to investigate processes and inefficiencies in the areas of planning and execution. Specific tools used include:
• Value-stream mapping
• Suppliers, inputs, process, outputs, customers (SIPOC) charts
• Responsible, accountable, consulted, and informed (RACI) charts
• Voice of the customer
• Fishbone diagrams
• “To Be” process workshops
• Data and capability analysis

Results

The project revealed significant opportunities for reducing end-to-end process cycle time, reducing human resource expenditures, and improving service quality by working in a more efficient and effective way. Improving these facets could deliver measurable savings toward CPS’s financial goal. One major finding was the acknowledgement and widespread discontent from the lack of any planning within the procurement process. This ultimately became the keystone for designing an improved process.

Central to the redefined process is a yearly cross-functional planning method that generates a written plan aligned with districtwide goals and budgets for the full fiscal year. The approved plan is then used to guide cross-functional team collaboration on process steps that were previously ad hoc. This maximizes the use of expertise in both the procurement and legal departments. The result is a new process with 40 percent fewer steps that promises to reduce cycle time by 50 percent for deliverables. It creates the mechanism to achieve the targeted expenditure reductions in a comprehensive and informed way.

The district

The Office of Procurement and Contracts at CPS supports the material sourcing needs for more than 680 schools throughout 234 square miles. With more than 40,000 full-time employees serving 404,000 students, the number of purchasers and the complexity and variety of items is immense. The highly variable socio-economic circumstances increase the diversity of procurement needs as well.

Figure 1: Diversity of students at Chicago Public Schools district

Figure 2: Diversity of teaching staff at Chicago Public Schools district

Demands on buyers

The diversity factors increase the workload for the Office of Procurement and Contracts. For example, its staff helps would-be purchasers (end users) to create solicitation documents that will accurately fill their needs and efficiently draw responses from qualified suppliers with competitive pricing. A trained and experienced team is needed to support the 40,000 CPS employees with such a complex activity as preparing the solicitation phase of the procurement process. This support can be extremely difficult, if not impractical, to provide when a process is ad hoc and lacks planning.

Financial need

The fiscal year (FY) 2013 budget for CPS was $5.8 billion with seven proposed expenditure categories (shown in figure 3).

Figure 3: Chicago Public Schools district’s proposed budget for FY 2013

Contracts are the second largest component in the school district’s budget (nearly 16%). Furthermore, the FY 2014 budget is projected to have a $1 billion shortfall in supporting revenue. Given these facts, improving the procurement process may be a significant means of realizing cost savings to help reduce the projected FY 2014 deficit.

Goal statement

CPS senior leaders identified the following goals as desirable outcomes from this project:
1. Reduce cycle time by 50 percent
2. Reduce defects by 50 percent
3. Increase customer satisfaction to 80 percent
4. Decrease the cost of procuring items by 50 percent.

A fifth goal was determined after the start of the project based on continued discussions with staff at multiple levels:

5. Improve data available for making approve/deny decisions on procurement requests

A structured approach

The decision to use a lean Six Sigma-structured approach hinged on two points:
1. A dedicated team at CPS had already completed a detailed value-stream map of the existing procurement process and highlighted opportunities for efficiency improvements (i.e., rework reduction) for several key deliverables.
2. From this value-stream mapping, the team envisioned segmenting the existing process into three distinct phases: plan, source, and validate. Plan and validate were new processes; source already existed.

The define, measure, analyze, improve, control (DMAIC) method is ideal for improving existing processes, and it was used to improve the source phase. The define, measure, analyze, design, verify (DMADV) method was employed to design the planning and validate phases. The plan phase was for districtwide use at the beginning of each calendar year, and validate would be a monthly process used throughout the year in conjunction with the Board of Education meetings.

Constraints

A main constraint on this project was the requirement to implement all changes using existing enterprise software. No IT funding was available. The new process had to tie into existing enterprise resource planning applications without any modification to them. This was a significant challenge for implementing an effective districtwide planning process. Because the existing procurement process used a custom Oracle application—the Electronic Board Report operating on a per-item basis—the new process could not be integrated without IT support. A bridging process from the output of the procurement cycle to the input of the purchasing activities had to be developed.

This highlights how critical it is to assess the impact of constraints early in the project planning: to plan for the design and implementation approach and to correctly assess the overall efficiency gains.

Figure 4: Making use of existing applications. Click here for larger image.

Tools employed

Various tools and techniques common to the DMAIC and DMADV methodologies were used during the course of the project.

Define

Critical to customer. The voice of the customer was obtained through structured interviews with key stakeholders using the procurement process. The following issues of concern were highlighted:
• The process should “work” for the people, not the other way around.
• Planning should drive the board reports.
• Clear communication of roles and responsibilities were needed.
• There were too many reactionary items as opposed to proactive items.
• Limit and reduce last-minute items.
• Items seeking board approval should be aligned with strategic goals.
• Board materials must have a calendar attached.
• There was no consistency of process or a clear set of rules that could be easily understood.
• “Should be” process should allow for seamless knowledge transfer.

The italicized words show a consistent theme: the need for additional structure and planning in the procurement process.

SIPOC. Suppliers, inputs, process, outputs, customers (SIPOC) workshops were held for each phase—plan, validate, and source—of the procurement process vision. For the plan phase, it was determined that the key inputs required tight coordination between the budget and procurement departments to provide guidelines, templates, and timelines for the sourcing and planning activity. The output of the process provided a detailed board approval plan coordinated with each department’s budget and aligned with the district goals.

Measure

Value-stream mapping. As noted earlier, the existing process was mapped in detail, and cycle-time issues with key deliverables were identified. Basic process parameters such as processing and cycle times were roughly estimated and are shown in table 1 (see end of article).

Items delivered from this activity include:
• Current state map including baseline metrics
• Improvement opportunities (e.g., kaizen events for “quick wins”)
• Future vision map and projected metrics
• Existing responsibility, authority, and accountability
• Workshop report

Customer satisfaction. A survey was conducted across the procurement and end-user departments that showed an overall customer satisfaction score of only 20 percent.

Analyze

Fishbone diagrams. One of the key process deliverables with the greatest amount of rework was creating the scope of goods and services for the supplier solicitation document. Two brainstorming sessions were held to better understand the root causes. The first session was held with appropriate staff from the procurement and legal departments; the second included staff from buyers and end users. See figures 5 and 6.

Figure 5: Fishbone diagram of issues when creating the scope of goods and services, as determined by the procurement and legal departments. Click here for larger image.

Figure 6: Causes and effects of issues when creating the scope of goods and services, as determined by buyers and end users. Click here for larger image.

All inputs were recorded on the fishbone diagrams verbatim during the two sessions, and later transferred to a spreadsheet for consolidation and analysis. A reduced set of causes was synthesized by grouping equivalent items together under a broader category, such as expertise. The resulting Pareto chart of the most common causes is shown in figure 7.

Figure 7: Pareto chart of root causes from fishbone diagrams. Click here for larger image.

Data review

Several decisions came from the reviewing the data:
• Changing the overall level of scope-writing expertise within a district so large was impractical, so existing expertise in the chief procurement office had to suffice. This led to modifying the existing source phase to include cross-functional, collaborative sessions as part of the documented process.
• Expectations and requirements would be addressed by publishing best-in-class examples of the scope of goods and services for various classes of items, and instructing the buyers to refer to these when collaborating with end users. Creating standard templates and checklists for novice writers were added to the list of project deliverables.
• Some root causes are not within the scope of the project to address, such as organization issues or complexity of requests, so time and effort should not be wasted on them.

Improve and design

Design. The plan and validate phases were designed to provide key functionality and output to support existing process steps:
• A commonly defined, districtwide plan in a written, centrally controlled document that guides the procurement process for the next five fiscal quarters
• Information critical for the board and CEO to make informed go/no-go decisions regarding procurement requests to meet ever-tightening budget demands
• A clearly defined timeline and justification guidelines for all procurement items
• A review process of the board approval plan (BAP) at the end of the plan phase that requires approvals from department heads, the board, and the CEO to be completed within a two-week time frame
• A bridging process used by the CEO office to create the required input documentation for the existing purchasing process

Improve. As can be seen from the high-level summary for the new process in figure 8, nearly every subphase of the source and validate phases had opportunity for efficiency improvement. The improvements included:
• Creating a standardized work-plan template for use by end users and buyers to manage the source and validate work once approval is granted, formatted for importing into Smartsheet as a collaborative PM template
• Improving guidance materials for performing market analysis
• Defining cross-functional meetings to support end users in writing the scope of goods and services, and executing the solicitation process
• Defining a “working BAP” to document any changes and progress for all approved items on the BAP

Control and verify

Significant effort went into the control aspect of this project, primarily in the form of process documentation and training. A SharePoint workspace was created for all process documents including training, the new process value-stream map, individual activity process documents, numerous templates, a working calendar, and both the BAP and the working BAP.

Value-stream map. This was created in Excel to allow easy scrolling from a web environment, with the longer-term goal of creating an interactive web (html) version that would work as a central dashboard, guiding users through each step of the process, identifying ownership and activity timing, and linking major activities to process control documentation.

Detailed control documents. Written in Microsoft Word, these are numbered to reference the specific activities in the value-stream map they support. This approach provides reference information now, but can easily be linked to the interactive value-stream map at a later date.

Training materials. These were done in Microsoft PowerPoint and used references to and screen shots from the other process documents.

Templates. These were created for the following value-stream mapping steps:
• Board approval plan (BAP)
• Working BAP: Defines additional information to be captured as the procurement process for approved items moves forward
• Scope of goods and services: Best-in-class examples for each procurement category to serve as templates, as well as a generic checklist for a “good” scope of goods and services
• Solicitations: Three different templates are used for bids—requests for proposals, requests for quotes, and requests for information
• Presubmittal presentation: Done in PowerPoint to clearly identify the key information to be presented at the supplier meeting as stakeholders prepare to respond to the solicitation
• Solicitation response evaluation scorecard: an Excel document that provides a simple checklist for the end user as he moves through the solicitation process
• Negotiation strategy template: A Microsoft Word document that consolidates key points, including demand estimates and target pricing
• Minority and women-owned business compliance: Documents district goals and how each procurement item contributes to those
• Recommendation letter: This is the formal mechanism for the chief procurement office, department head, and award-evaluation committee to document that all required legal processes were followed and they approve the award decision.

The final design

A high-level description of the redesigned process is shown in figure 8.

Figure 8: Final redesigned procurement process. Click here for larger image.

Because the new process was implemented concurrent with the project’s conclusion, no data are yet available for the key process attributes such as cycle time and processing time. One complete planning and procurement cycle, a full fiscal year, is required as the minimum data set to begin quantifying efficiency improvement. However, there were several accomplishments.

Process steps. Procurement process steps were reduced by 54 percent.

Process map. A process map was created that documents the entire process in detail, including activity ownership, relationship, and time frames.

Collaboration. Cross-functional meetings were designed into the process, offering the following advantages:
• Improved leveraging of limited expertise (buyers, attorneys)
• Earlier involvement of legal department
• Reduction of rework cycles stemming from serial review of key steps such as writing the scope of goods and services, and the solicitation document

Documentation. Detailed controlled documents, templates, checklists, and training materials represent a level of integrated process documentation not previously seen at CPS, and will support rollout of the new process and sustain the required culture change.

Metrics. A complete set of process metrics were identified with a phased approach of implementation suggested. The minimum set are those listed in table 1, cycle time and processing time. If properly used, data for both should be available from the new process tools, the working BAP, and the standard work-plan template.

Table 1: Key process attributes—existing vs. new. Click here for larger image.

Continuous improvement. The items above create a strong foundation for long-term adherence to the new process (control), which can be expanded (continuous improvement) as resources allow, including:
• Implementing interactive links between value-stream maps and detailed control documents
• Implementing a full suite of identified key process indicators
• Automating process flow using SharePoint workflows
• Enhancing project management, data collection, and key process indicator assessment capability by incorporating the standardized work-plan template into a collaboration tool such as Smartsheets.

Conclusions

Results of the project show there is significant opportunity to improve nonmanufacturing processes by applying lean and Six Sigma methods, even when budget and IT support are severely constrained.

Most organizations already have the key understanding of what isn’t working. Staff is eager to share this knowledge and brainstorm solutions, and can be very effective when brought together in a structured setting with well-defined methods and goals. Minimal enterprise applications are typically available that can be leveraged to successfully implement the results of the DMAIC and DMADV projects. This can be achieved with a minimal commitment in staff resources and a dedicated project leader knowledgeable in Six Sigma techniques.

Acknowledgements
The authors thank the participation of the cross-functional core team. These individuals were committed to the team while maintaining their dedicated responsibilities. In addition, the authors acknowledge the support of the CPS project sponsors, which included individuals from the Chicago Board of Education, CPS CEO’s office, and CPS senior leadership team. Their continued support was instrumental to the project’s success.

About The Authors

George Chemers

George Chemers is a lean Six Sigma consultant. Previously he was the quality/continuous improvement director at Motorola in charge of major change initiatives, improving internal processes, reducing cost-of-quality expenditures, and understanding customer needs and expectations. He is also a certified Six Sigma Black Belt and certified ISO 9001 auditor.

Mike Moutrie

Mike Moutrie is a quality, process, and product improvement professional and management consultant. He has extensive experience in quality management systems, process design, lean Six Sigma, product life-cycle management, and statistical process control methods. At Motorola Moutrie he managed teams to develop, launch, and sustain complex consumer electronics products.

Kimberly Parpala

Kimberly Parpala is the director of project management/engineering tools at Motorola Mobility (a Google Co.), where she assesses technology and customer requirements, turning them into solutions that increase productivity. Her background includes business development, lean Six Sigma, change management, training/communication, and quality control systems.