Application of Dependency Structure Matrix in Time Scheduling

From apppm
(Difference between revisions)
Jump to: navigation, search
(Background)
(How to Read the Matrix)
Line 8: Line 8:
  
 
=How to Read the Matrix=
 
=How to Read the Matrix=
 +
 +
The matrix consists of all the project tasks listed as both rows and columns. This means that the matrix is an nxn square matrix where n is the number of activities. The diagonal of the matrix lists the elements themselves and the off-diagonal elements lists the dependencies by either and ‘X’ mark or a ‘0’/left out. It is crucial for the use of the activity based matrix that all the tasks are listed in progressing order from the first in the upper left to the last in the bottom right.
 +
There are two ways of reading the matrix, the IR/FAD and IC/FBD convention. The first one is the one typically used for scheduling activities in construction. This states that the rows list all the requirements or inputs that a certain task needs to have to begin. The columns on the other hand list all the outputs from the activity. This means that all the interdependencies can be stated for the whole project in a single matrix.
 +
In the other convention, the way of looking at the rows and column is reversed effectively meaning that the matrix is transposed. Thus, all the observations in the following are based on the IR/FAD convention but can easily be converted to the IC/FBD convention by reversing the rows and columns. Especially the elements in the upper triangle of the matrix are important. In the IC/FBD convention these elements are therefore in the bottom triangle.
  
 
=Using the Matrix=
 
=Using the Matrix=

Revision as of 13:35, 9 June 2017

In planning management with regards to time scheduling there are several ways to get an overview of the number of activities in a timeline like the Gantt plot or using the Critical Path Method (CPM) to estimate the time used and needed for activities dependent on each other. But CPM is limited in the way that it only looks at sequential activities though it is possible for a design case to have both dependent activities as well as interdependent activities. This means that the information flow is not necessarily linear but can be cyclic giving rise to situations where some activity is initiated but still needs information from a later activity. The structure of these activities through time can be sorted in a Dependency Structure Matrix (DSM) which marks all the interdependencies in a square matrix. The matrix can both be used for visualization and optimization of information flows. This article will discuss the application of DSM for use in a time scheduling situation of project management. Here some of the common ways to rearrange or change the matrix to reduce loops of information flow or to let non-dependent activities run in parallel will be discussed.

Contents

Background

For planning in construction of buildings a simple Gantt chart may be of use to schedule the work flow of different tasks. Otherwise a CPM analysis (Critical Path Method) or PERT model () can be used to sequence the different tasks such that the total cost and work time is optimized. But these models are limited to progressive flows of work meaning that the completion of one task leads to the possible start of another activity. But when designing the building a lot of iterative procedures are performed which cannot be modelled by CPM for instance. These iterative procedures imply that some tasks are not just dependent on each other but also interdependent. These cycles of iteration can lead to over-designing of the project which entails waste of time and extra costs. Donald V. Steward proposed a method in 1981 using a Design Structure Matrix to optimize the information flow for the design scenario. It was later called Dependency Structure Matrix as it went from being useful mostly for the information flow during the design phase to also being applicable for other cases, not only scheduling. Several types of DSM can be used for different purposes. There are ………… but this article will focus on the use of activity based DSM.

How to Read the Matrix

The matrix consists of all the project tasks listed as both rows and columns. This means that the matrix is an nxn square matrix where n is the number of activities. The diagonal of the matrix lists the elements themselves and the off-diagonal elements lists the dependencies by either and ‘X’ mark or a ‘0’/left out. It is crucial for the use of the activity based matrix that all the tasks are listed in progressing order from the first in the upper left to the last in the bottom right. There are two ways of reading the matrix, the IR/FAD and IC/FBD convention. The first one is the one typically used for scheduling activities in construction. This states that the rows list all the requirements or inputs that a certain task needs to have to begin. The columns on the other hand list all the outputs from the activity. This means that all the interdependencies can be stated for the whole project in a single matrix. In the other convention, the way of looking at the rows and column is reversed effectively meaning that the matrix is transposed. Thus, all the observations in the following are based on the IR/FAD convention but can easily be converted to the IC/FBD convention by reversing the rows and columns. Especially the elements in the upper triangle of the matrix are important. In the IC/FBD convention these elements are therefore in the bottom triangle.

Using the Matrix

Shearing

tearing

Advantages of the Method

Disadvantages of the Method

Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox