Wednesday, August 21, 2013

Top-Down and Bottom Up Processes

The process of starting with world objects and modeling using entity-relationship diagrams is referred to as a top-down process. Starting with one large table and functional dependencies using normalization is referred to as bottom-up development.

What are the advantages and disadvantages of each method? Are there any inherit dangers with either method? Which would you prefer to use? Is it really an either-or situation?

Top-Down and Bottom Up Processes

Top Down
Top-Down is deductive reasoning. It can be used in conjunctions with analysis and decomposition. Breaking down a system to gain insight into different elements is the top-down approach. First a total system is developed, and then subsystems are detailed.
There may be many different levels until everything is reduced to a whole. To put it in simple terms, top-down approaches start with the big picture. This concept is broken down into smaller segments for ease of understanding and learning.

In business top down can be illustrated by leadership, business integration, processes, learning and data. Leadership requires an excitement level about improvement processes. Leaders believe that there are processes that can improve the company. Business integration is communication about the new initiative. Processes designate a small team that begins to map current processes from the top to the bottom.  Learning is the process where training in the improvement issues are presented to the entire company. Data gathering at this point in the process is just beginning.

An advantage of top-down reasoning includes the clear statement from executive. Training is given to numbers of employees and a communication is built. There is a full list of process and documentation is required.

Challenge is time. There are many activities, but the analyzing and improving processes are slow yet thorough. Training gives the tools and concepts, but not actual work. Basically there is little change with top down reasoning.

Bottom Up
Bottom up is inductive reasoning. The original idea is the sub-set of the emerging system. This is an information process based on data coming in from difference sources to form a perception. Information processing is determined on incoming data from outside and. individual base elements are described in great detail.

All systems are linked until a complete system is formed. Think of bottom-up as a seed. The beginnings are small but they grow in complexity to form the whole.
Leadership is connected and engages employees in improvement efforts. 
Managers determine what processes in their department needs to be fixed and assign employees to "fix" the processes. Most processes are narrow with improvements implemented in only one area. The learning is hands on training either from a manager, vendor, or professional training organization. Data is real-time and the problems are known up front.
As the process improves the team knows the results, but actual data is limited.

Utilizing the bottom-up approach, changes can be researched and implemented immediately.  Efforts are focused on specific defects and problems and learning is focused on real work problems.
Small improvement in processes and thinking are made quickly, but widespread changes take quite a bit of time to be recognized and utilized.

There is no strategy for focus and improvements generally do not have a huge impact on the entire company. Improvement scan spread, but they can also die quickly. Managers can only fix their own departments (Sweet, 2011).

There is no right or wrong method in reasoning. If your company wants changes in processes to be slow and steady use top-bottom reasoning. If you need quick fixes in departments, bottom-up may be the best approach. Mix the two processes together to come up with the real answers.

Sweet, Shelly (2011). Which is Best for Us? Top-Down, Bottom-Up or Middle Out? Available: Last accessed: 3 April 2013.

Elad Shalom,
Senior Consultant at SwiftRadius

Developing Accurate Cost Estimates

What is the Biggest Problem in Developing Accurate Cost Estimations? Why?

Cost Estimation Methods 

Developing the estimated cost of a project, can be the variance between completing projects on time and being able to complete the project on budget. Techniques for cost estimations are very important and particularly if you are the project manager. Estimating what processes cost and how they work to provide a finished product should be a part of the project proposal. Study the techniques that that you feel will give you the most accurate estimation method.

Types of Cost Estimation Methods
  • Analogous estimating is learning from precedents.  Read though past projects and determine how the cost estimating was based. Analogous estimating can provide a continuous basis for developing estimates based on past learning. Project parameters that can be estimated include costs, budget, scope, and duration. Use analogous estimation to determine the complexity plus the size of the entire project. Compare a current activity to that of an activity that finished in the recent past when the information about the current project is unclear or unavailable. Using expert judgment is highly reliable (Project Management Knowledge, 2010).
  • Parametric estimating is a very accurate method of determining costs. This is based on a previous cost model. Using the cost per line of code, square foot or per cubic inch is parametric estimating. Project managers use this method for construction and certain type of software development. Verified cases provide the basis for parametric estimating. When used correctly, parametric cost estimating is highly accurate. However, there are numerous challenges, in this type of cost estimating and adverse effects can be the result if the formulas and premises are not used properly. Parametric cost estimating is very popular in software development to estimate the development and product life cycle costs.
  • Program Evaluation and Review Technique (PERT) identifies three separate estimates first based on the optimist outcome and then based on the pessimistic outcome. This third point is the premise that the larger the distance between pessimistic (worst case scenario) and optimistic (best case scenario) values the less like the project will succeed. The distance is calculated: O+(4*M)+P/6.  PERT separates each section of the project into events and activities and schedules them in their proper sequence. Paths connect each event and the critical path is the connecting point duration of the critical path is how long the project is estimated to take. There are always delays factored into the paths. However, this method does not find the best way to complete a project (Mind Tools, 2012).
  • Rule of thumb cost estimating is a universally accepted estimating basis. Rule of thumb estimates are unique to every project and to each industry. Rule of thumb cost estimating looks at several completed projects and reviews them as a benchmark of measurement and cost estimating.  According to the Association for the Advancement of Cost Engineering International a cost estimate is "an evaluation of all the costs of the elements of a project or efforts as defined by an agreed-upon scope" (US Army Corp of Engineers, 2000). Rule of thumb is the total estimated cost of a project and is dependent on how well the project is actually defined or what the scope of the project is.

What is the Best Cost Estimating Method?

Every method has its own complications and inherent problems. However, when using analogous or precedent cost estimating, there is a basis and a foundation. All projects have the same elements of supplies, time, budget and finish. By following models that have been already proven, an accurate method of estimating costs can be reached.

Farens, Daniel V. (1999). Parametric Estimating – Past, Present and Future. Available: Last accessed: 22 March 2013.
Project Management Knowledge (2010). The Ultimate Resource for project managers: Analogous Estimating Techniques. Available: Last accessed: 22 March 2013.
U.S. Army Corps of Engineers (2000). A Guide to Developing and Documenting Cost Estimates During the Feasibility Study. Available: Last accessed: 22 March 2013).

Elad Shalom,
Senior Consultant at SwiftRadius