Applying the Software Development Process Improvement Model

< Free Open Study >

Applying the Software Development Process Improvement Model A 50-person software development organization is our example of applying the six-step software process development model. For purposes of this example, the company will be called Atlanta Elevator Company (AEC). AEC has three years of project metrics in its database and is performing at CMM Level 3 maturity. It is a standalone business unit of a major elevator hardware manufacturer that develops software only to generate revenue. There are no product sales to cloud the revenue stream. Full financials have been analyzed to arrive at the revenue numbers used for costing. In the last three years of project record-keeping, the organization billed $15,000,000 at an average hourly rate of $165.00. The total number of billed hours was 90,900. AEC uses a timesheet system to track all billable and nonbillable work broken out into projects. Each project has a series of phases that also can be tracked. AEC wants to begin a continuous process improvement initiative and will use the current set of metrics to baseline the entire organization and look for waste targets of opportunity to begin the improvement process. This is not a cost-accounting exercise. The continuous process improvement initiative must look at AEC from the customer's perspective. Areas that may need more business financial information are recommended to the internal business practices re-engineering task force. Plan AEC needs to baseline the software development process over the last three years. There is no established categorization of value- versus non�value-added tasks done on a project-by-project basis or organization-wide. This project will set that baseline through the use of a value-analysis worksheet. This worksheet will cover the past three years of data. The adapt phase will focus on recommending changes to the current metrics system to collect information not currently gathered for completing customer-based value analysis. The improve phase will implement those collection changes. The AEC project manager, Mr. Lifté, will develop estimates based on past history with process improvement efforts. There are a minimum number of internal resources that can assist on these projects. They are reserved for software developers between projects who have a couple of days or, at most, a week to commit to an internal project. See Figure 26-11. Figure 26-11. AEC Process Improvement Plan Phase Worksheet Research The basic AEC information needed for the value analysis was available in two easy-to-access databases. The systems admin merely directed Mr. Lifté to the correct share and charged nothing to the process improvement project because the effort took less than five minutes. The information required has been collected in a consistent format for the past three years. It requires no special access, and it looks like there will be little modification needed in the next phase. See Figure 26-12. Figure 26-12. AEC Process Improvement Research Phase Worksheet Observe The project manager copied all of the data found in the research phase into a separate database to be used for analysis. This took a total of five hours. There were no problems in accessing the data. The financial information needed was on a report generated monthly for the organization business manager. One of the interesting occurrences of data collection for process improvement is that there exists a lot of data that is not being used. Even in relatively young organizations, data collections activities, once begun, never seem to end. Megabytes of data are there just for the asking. Waste is not that difficult to find. See Figure 26-13. Figure 26-13. AEC Process Improvement Observe Phase Worksheet Analyze The analyze phase took the database developed in the previous phase and did simple statistical analyses using Microsoft Excel. Using the value-analysis template (Table 26-2), the information developed from the AEC metrics data was entered into the template. Figure 26-14 shows the process improvement phase worksheet for the analyze phase. Note that the estimate was 16 hours, but the actual was 20. The data analysis required the project manager to use four hours of a financial analyst's time to complete the statistical study. Figure 26-14. AEC Process Improvement Analyze Phase Worksheet Table 26-4 shows the results of the analysis. These results will be further analyzed in the adapt phase to produce process improvement recommendations. Adapt Based on the three-year value analysis of Table 26-4, a set of information visualization graphics needs to be prepared to lay the foundation for the adaptation recommendations. Figure 26-15 shows a gross distribution of the value-analysis data. This chart represents the total hours recorded in the metrics system for the past three years: Value-added� What the customer really wants and sees as quality�69,300 hours Essential� What the customer will accept as necessary, but is not willing to really pay for�18,025 hours Nonessential� What the customer will readily admit is overhead�83,700 hours Hours billed� What the customer really paid for�90,900 hours Figure 26-15. Gross Distribution from Value Analysis Analyzing how the hours were broken out in the value analysis versus what was paid for by the customer, this pattern emerges: The customer paid for all value-added work. The customer paid for all non�value-added essential work. The customer paid for 3,575 hours of non-value added, nonessential hours. The customer paid for 131% of what is considered value work. Table 26-4. AEC Three-Year Value Analysis Project Identifier: AEC Baseline Historic Value Analysis Value-Added Non�Value-Added Essential Nonessential Design 14250 Acceptance Testing 7200 Bottlenecks   Documentation 2500 Moving Data Between Steps   Correction 700 Implementation 28000 Planning 9250 Delays   Installation 850 Setup 200 Error Analysis   Requirements 19500 Some Process 625 Expediting   Shipping 3000 Training 750 Extra Paperwork   Upgrading 1200     Extra Unwanted Features   Value-Added Non�Value-Added Essential Nonessential         Installing Software Tools 11000         Management 62500         Metrics           Modification           Process Improvement 9500         Recall           Retest           Rework   Total Value Added: 69300 Total Essential: 18025 Total Nonessential: 83700 Project Value to Total: 40.5%         Percent Essential to Nonessential: 21.5%     Percent Nonessential to Total: 48.9% Percent Value Plus Essential to Total: 51.1%         Going back to the purpose of this first, the organization-wide process improvement project was intended to define the baseline for continuous process improvement. This baseline has now been set, and the following questions arise for further analysis: If this organization's customers are billed and are paying for hours that are traditionally viewed as non�value-added, what is their satisfaction level with the product received? Forty-eight percent of all hours recorded for the past three years were non�value-added, nonessential. Figure 26-16 shows the distribution of where those hours were recorded. The easy target is the enormous chunk of 62,500 hours in management. What is really in this bucket, and how can the waste be driven out? Figure 26-16. Non�Value-Added, Nonessential Hours Distribution Looking again at the nonessential portion of the value analysis, why are there no recorded values for the categories bottlenecks, delays, error analysis, expediting, extra paperwork, extra unwanted features, metrics, modification, recall, retest, and rework? Using all the information presented for AEC, it is left to the reader and student to complete the recommendations in Table 26-5. This should be done as either an individual homework assignment or an in-class team exercise. See Figure 26-17. Figure 26-17. AEC Process Improvement Adapt Phase Worksheet Improve Based on the individual or teamwork done in the previous section, complete Figure 26-18 for the improve phase. This can be done as individual homework or in-class team exercises. Figure 26-18. AEC Process Improvement Improve Phase Worksheet Table 26-5. Adaptation Recommendation ID Description Implementation Risk Cost Benefit 1 Customer Satisfaction Analysis     2 Waste in Management     3 Missing Metrics Data     4       5

< Free Open Study >