Software Problems: Summary

The following learning objectives are covered in this lesson:

• Apply a generic problem-solving model to an acquisition situation.
• Apply one or more selected qualitative tools (e.g., fishbone diagram) to resolve a problem.
• Identify developer practices essential for creation of high quality software.
• Identify the requirements for interoperability testing.

1.    One problem-solving technique is the cause and effect diagram or "fishbone" diagram.  By analyzing all the possible causes of a problem, the fishbone diagram focuses on determining the root cause of a problem, rather than on symptoms or solutions.  Typically, the fishbone diagram begins with a statement of the problem in a box on the right side of the diagram--the "head" of the fish.  Then categories of major causes are identified and drawn to the left--the "bones" of the fish.  These major causes are broken down into all the related causal factors that might contribute to the major causes.  Finally, the causal factors are examined and narrowed down to the most significant elements of the problem to determine the ultimate cause or causes.

2. The Software Program Managers Network has identified several software best practices based on interviews with software experts and industry leaders.  Here is a synthesized list of some of those characteristics, which are essential for the creation of high quality software: Adopt Continuous Program Risk Management

Risk management is a continuous process beginning with the definition of the concept and ending with system retirement. Risks need to be identified and managed across the life of the program.

Estimate Cost and Schedule Empirically

Initial software estimates and schedules should be looked on as high risk due to the lack of definitive information available at the time they are defined.

Use Metrics to Manage

All programs should have in place a continuous metrics program to monitor issues and determine the likelihood of risks occurring. Metrics information should be used as one of the primary inputs for program decisions.

Track Earned Value

Earned value requires each task to have both entry and exit criteria and a step to validate that these criteria have been met prior to the award of the credit. Earned value credit is binary with zero percent being given before task completion and 100 percent when completion is validated.

Track Defects against Quality Targets

All programs need to have pre-negotiated quality targets, which is an absolute requirement to be met prior to acceptance by the customer. Programs should implement practices to find defects early in the process and as close in time to creation of the defect as possible and should manage this defect rate against the quality targets. Meeting quality targets should be a subject at every major program review.

Treat People as the Most Important Resource

A primary program focus should be staffing positions with qualified personnel and retaining this staff through the life of the project. The program should not implement practices (e.g., excessive unpaid overtime) that will force voluntary staff turnover. The effectiveness and morale of the staff should be a factor in rewarding management.

Adopt Life Cycle Configuration Management

All programs, irrespective of size, need to manage information through a preplanned configuration management (CM) process. This discipline requires as a minimum:

• Control of shared information
• Control of changes
• Version control
• Identification of the status of controlled items(e.g., memos, schedules) and
• Reviews and audits of controlled items.

Manage and Trace Requirements

Before any design is initiated, requirements for that segment of the software need to be agreed to. Requirements need to be continuously traced from the user requirement to the lowest level software component.

Use System-Based Software Design

All methods used to define system architecture and software design should be documented in the system engineering management plan and software development plan and be frequently and regularly evaluated through audits conducted by an independent program organization.

Ensure Data and Database Interoperability

All data and database implementation decisions should consider interoperability issues and, as interoperability factors change, these decisions should be revisited.

Define and Control Interfaces

Before completion of system-level requirements, a complete inventory of all external interfaces needs to be completed. Internal interfaces should be defined as part of the design process. All interfaces should be agreed upon and individually tested.

Design Twice, Code Once

Traceability needs to be maintained through the design and verified as part of the inspection process. Design can be incrementally specified when an incremental release or evolution life cycle model is used provided the CM process is adequate to support control of incremental designs.

Assess Reuse Risks and Costs

The use of reuse components, COTS (Commercial Off-The-Shelf), GOTS (Government Off-The-Shelf) or any other non-developmental items (NDI) should be a primary goal, but treat any use as a risk and manage it through risk management.

Inspect Requirements and Design

All products that are placed under CM and are used as a basis for subsequent development need to be subjected to a formal inspection defined in the software development plan. The program needs to fund inspections and track rework savings.

Manage Testing as a Continuous Process

All testing should follow a preplanned process, which is agreed to and funded. Every test should be described in traceable procedures and have pass-fail criteria.

Compile and Smoke Test Frequently

Smoke testing should qualify new capability or component only after successful regression test completion. All smoke tests should be based on a traceable procedure and run by an independent organization (not the engineers who produced it). Smoke test results should be visible and provided to all project personnel.

3.    Interoperability problems can best be identified through the use of actual, live systems to mitigate risk.  Joint interoperability is defined as the ability of systems to provide services to and accept services from other systems and to use the services exchanged to enable them to operate effectively together.  The Joint Interoperability Test Command is responsible for verifying the interoperability of systems to the parameters outlined in the ICD, CDD, CPD and ISP.