Metrics Program Part - 2

Standish Group Chaos Report - 26% Project are completed on time and budget

28% Projects scrapped before actual end

46% - In between the above 2 i.e. schedule or cost overrun.

A delay of 3 moths in IT project has revenue loss of 20%, as its time to profit has reduced and corresponding release delay.

Operational Metrics : Which help us understand benefit calculation :

1. Time to Market

2. Time to Profit

3. Cost of Non-Quality

4. Cycle time to single process

5. Cycle time to defect correction

6. Resue Rate

Productivity Metrics :

1. Cost \ engineer

2. Leraning curve Cost

3. Cost\ employee\country

4. Cost Evolution

5. Output rates\engineer

6. Cost\seat

E3 - Extract, Evaluate and Execute

While reading the book "

Best practices in software measurement:

how to use metrics to improve ...

By Christof Ebert, Reiner Dumke, Manfred Bundschuh"it talks of 3 steps to be followed in metrics, it suddenly striked me it can be applied to any activity as part of daily chores.

Decision making - well thats what metrics are needed for

1. Extract - Info, metric required for

2. Evaluate - The info needs to be evaluated for specific status.

3. Execute : Rather Act, take decision based on the above steps.

Program Management vis Project Management

Program Manager :

1. He keeps a eye on the triple constraints : time , cost and scope of the projects within his program with additional emphasis on business targets.

2. Program Manager Decisions are to meet tactical and strategic goals of the company.

Project Manager :

1. He keeps a eye on the triple constraints : time , cost and scope of the project.

2. He is more delivery and execution focussed.

Introduction to DICOM -

What is DICOM :

DICOM stands for - Digital Imaging and Communication in Imaging.

A medical standard which has evolved over last 20 yrs. It is a one stop shop for digital medical imaging (i.e. data transfer,storage and display ). All DICOM compliant companies have a PACS - Picture Archiving and Communication Systems.

PACS comprises of 3 main parts :

1. Image acquisition systems - called modalities in Medical Domain (MRI, CT, Ultrasound).

2. Digital Image Storage devices

3. Viewer Application s : Desktops on which acquired images are stored- Workstations.

PACS implements DICOM Standard :

Advantages of DICOM :

1. Universal standard in digital Imaging.

2. It supports 16bit of Greytones(around 65,000) in imaging, thus helping capture nuances of medical image, in comparison Jpeg or BMP can have only 256 shades of Greytones.

3. Storage of numerous image related parameters with the image.

4. 2000+ standardised medical attributes(Data dictionary), which helps in accurate diagnostics.

Managing Multiple Projects - Dos & Donts

How many times have you faced with a situation of handling more than 1 project- say 5-8, wow are we talking about Program Management. Well welcome to real world - well we do come across project managers handling 5-10 projects. Rather than being frustrated and fatigued, which frankly i am, handling 5 projects , lets see how we can do the juggling act.

Handlling multiple projects with shared resources and no dedicated teams, is indeed a hell of a challenge.

Dos

Project Tracking : Status meeting at least twice a week, with MoM circulated monitor projects based on the MoM and the agenda.

Know Priority of the Projects :

Know which projects are critical, and any schedule slip can impact land you in soup.

Dont`s

Overallocation : Resources in shared projects if over allocated, no project deadlines are going to be met.

e.g Resource A, has load factor 80% i.e can finish 4 days task in 5 days\ week.

If allocated on project A- 50% - would work of 2.5 days in 3.1 calender days.

Resource A can now be allocated for 30% i.e. can work 1.5 days in 1.9 calender days in the same week

In a Scenario - where Project manager B books the resource for 50% in same week, he would be over allocated and either of your of Project Manger B project would be impacted.

If using Project Server : check Resource allocation to confirm if the resource is not overallocated.

Few Intresting Articles :

http://articles.techrepublic.com.com/5100-10878_11-1061892.html

Vision to Process

Policy - a direction; a writing that defines a business objective, plan or course of action.

Procedure - the what; a writing that defines a business process and assigns responsibility.

Process - the how-to; a writing how a specific operation is to be performed.

Goal is a position or place which an organization wants to attain. Goal is close to reality and is achievable by people.

Objective is the plan, to make and execute particular actions which must be taken to finish the gap between the current state and the desired one.

Mission is the method of defining objectives and making strategies to get those objectives.

Vision includes the foresightedness which is required to get the essential skill to organize existing resources, to achieve the desired outcome.

Company first make a Vision, then mission, in order to accomplish mission they make objectives and at the end they make goals (end result/place which organization want to get).

To take the Analogy little further to get the objective, Policy are framed, the what of Policy is Procedure and Policy is how the procedure is executed.

Metrics Program

Having worked for a CMM 5 Org. which is known in the industry for its processes and quality metrics, this effort is to collate Metrics which can help us to scrutinise projects with right perspective.

Defect Metrics

Defect Removal Efficiency (DRE)
DRE is a measure to detect defects before delivery.It is calculated as a percentage of the defects identified and corrected internally with respect to the total defects in the complete project life cycle. Thus, DRE is the percentage of bugs elimiated by reviews, inspections, tests etc.
Here is an interesting Defect Removal Efficiency calculator by Harvey Roy Divinagracia:http://www.engin.umd.umich.edu/CIS/course.des/cis525/java/f00/harvey/DRE_Calc.htm

Defect Removal Efficiency where - E1 - Defects found before IT & E2 - Total Defects till date

Project/Activity: XXX DRE = 0.32070708 (32.07071 %) where E1 = 127 , E2 = 269.

Industry Standard - 85%

According to Software Productivity Research chairman Capers Jones’ paper, “Software Defect-Removal Efficiency” (IEEE Computer, April 1996), software products can be considered “high quality” if the defect removal efficiency reaches 95 percent. The defect removal efficiency is a quality metric calculated by dividing the development defects (defects reported prior to product release) by the total number of defects (the sum of the development defects plus the defects reported during the first year of product use). In other words, for best-in-class companies, the investment in defect removal yields 95 percent of the defects prior to product release—and according to Jones, the industry standard of 85 percent just isn’t good enough.

Defects\KLOC

Number of defects found per 1000 Lines of Code.
With a Japanese client that i worked they expected 0.25defects \KLOC i.e 1 defect\KLOC.

Here's an interesting blog post that cites some figures from books: http://amartester.blogspot.com/2007/04/bugs-per-lines-of-code.html

The book "Code Complete" by Steve McConnell has a brief section about error expectations. He basically says that the range of possibilities can be as follows:
(a) Industry Average: "about 15 - 50 errors per 1000 lines of delivered code." He further says this is usually representative of code that has some level of structured programming behind it, but probably includes a mix of coding techniques.
(b) Microsoft Applications: "about 10 - 20 defects per 1000 lines of code during in-house testing, and 0.5 defect per KLOC (KLOC IS CALLED AS 1000 lines of code) in released product (Moore 1992)." He attributes this to a combination of code-reading techniques and independent testing (discussed further in another chapter of his book).
(c) "Harlan Mills pioneered 'cleanroom development', a technique that has been able to achieve rates as low as 3 defects per 1000 lines of code during in-house testing and 0.1 defect per 1000 lines of code in released product (Cobb and Mills 1990). A few projects - for example, the space-shuttle software - have achieved a level of 0 defects in 500,000 lines of code using a system of format development methods, peer reviews, and statistical testing."

The NASA has a defect density of 0.004 bugs/KLOC (vs 5.00 bugs/KLOC for the Industry) but this has a cost of $850.0/LOC (vs $5.0/LOC for the Industry). Source: agileindia.org/agilecoimbatore07/presentations/… – Pascal Thivent

Defect Density - Defects\KLOC is a flavor DD

Defect-Density (DD) Measures the number of defects in a perticular size of code. it is measured as follows:

DD = Defects/Size of the Project 

Size of Project can be Function Points, Feature Points, Use Cases, KLOC etc

Defect Age :

http://www.softwaretestingfundamentals.com/2009/05/defect-age.html

Defect Age (in Time) is the difference in time between the date a defect is detected and the current date (if the defect is still open) or the date the defect was fixed (if the defect is already fixed).

Defect Age (in Phases) is the difference in phases between the defect injection phase and the defect detection phase.

This article explains Defect age in Time and by Phase. I have used with Phase and helps to identify the problem area. Though most of time we are tempted to conclude Phase Injected is Coding. Deeper probing reveals its the requirement phase which should have been the injection phase.

The latter phase they are identified the higher the efforts to fix the defects and more chances of the 3 sides of traingle impacted, Cost, Time and schedule.

Number the phases from Requirements as 1 to Acceptance to 8, the lower the number the better the project health.

Average Defect Age = (Activity Detected - Activity Introduced)/Number of Defects

Effort Metrics :

What was the accuracy of estimating the actual value of project effort?

Effort Estimation Accuracy (EEA) - Motorola uses this

What is Rookie Ratio, Bulge Mix

Rookie Ratio is defined as number of people with less than 2 years employment. Recently employed people are less stable than old. They are also less efficient, because they have not yet socialized into the tradition of the organization, so they do not know the most efficient way around.

Bulge mix is calculated as a percentage of employees with experience below three years over the total number of billable employees.