- … have long durations<\/li>
- … have a high amount work<\/li>
- … have little slack<\/li>
- … have a high criticality<\/li><\/ul>\n\n\n\n
The blog post shall also encourage readers to think outside of the possibilities that established project management software offers and to define for yourself<\/em> which tasks are critical or relevant for you and how this information shall be determined.<\/p>\n\n\n\n
Preconditions<\/h2>\n\n\n\n
In order to use the approach described here, you should:<\/p>\n\n\n\n
- … have access to a Linux machine or account<\/li>
- … have a MySQL<\/a> or MariaDB<\/a> database server set up, running, and have access to it<\/li>
- … have the package graphviz<\/a> installed<\/li>
- … have some basic knowledge of how to operate in a Linux environment and some basic understanding of shell scripts<\/li><\/ul>\n\n\n\n
Description and Usage<\/h2>\n\n\n\n
The base for the colored network graphs is a plan in Microsoft\u00ae<\/sup> Project<\/strong> which needs to be stored as XML<\/strong> file. The XML file is readable with a text editor (e.g., UltraEdit<\/a>) and follows the syntax in [1<\/a>]. The XML file is then parsed, the relevant information is captured and is stored in a database (Step 1). Then, selected information is retrieved from the database and written to a script in dot<\/a><\/em> language which subsequently is transformed into a graph using the package graphviz<\/a> (Step 2). Storing the data in a database rather than in variables allows us to use all the computational methods which the database server offers in order to select, aggregate, and combine data from the project plan. It also allows us to implement Step 2 in PHP so that graphs can be generated and displayed within a web site; this could be beneficial for a web-based project information system.<\/p>\n\n\n\n
Step 0: Setting up the Database<\/h3>\n\n\n\n
At the beginning, a database needs to be set up in the database server. For this, you will probably need administrative access to the database server. The following SQL script sets up the database and grants access rights to the user gabriel<\/em>. Of course, you need to adapt this to your needs on the system you work on.<\/p>\n\n\n\n
# Delete existing databases\nREVOKE ALL ON projects.* FROM 'gabriel';\nDROP DATABASE projects;\n\n# Create a new database\nCREATE DATABASE projects;\nGRANT ALL ON projects.* TO 'gabriel';\nUSE projects;\n\nCREATE TABLE projects (proj_uid SMALLINT UNSIGNED NOT NULL AUTO_INCREMENT PRIMARY KEY,\\\n proj_name VARCHAR(80)) ENGINE=MyISAM DEFAULT CHARSET=utf8;\nCREATE TABLE tasks (proj_uid SMALLINT UNSIGNED NOT NULL,\\\n task_uid SMALLINT UNSIGNED NOT NULL,\\\n task_name VARCHAR(150),\\\n is_summary_task BOOLEAN DEFAULT FALSE NOT NULL,\\\n is_milestone BOOLEAN DEFAULT FALSE NOT NULL,\\\n is_critical BOOLEAN DEFAULT FALSE NOT NULL,\\\n task_start DATETIME NOT NULL,\\\n task_finish DATETIME NOT NULL,\\\n task_duration INT UNSIGNED DEFAULT 0,\\\n task_work INT UNSIGNED DEFAULT 0,\\\n task_slack INT DEFAULT 0,\\\n rem_duration INT UNSIGNED DEFAULT 0,\\\n rem_work INT UNSIGNED DEFAULT 0,\\\n `%_complete` SMALLINT UNSIGNED DEFAULT 0,\\\n PRIMARY KEY (proj_uid, task_uid)) ENGINE=MyISAM DEFAULT CHARSET=utf8;\nCREATE TABLE task_links (proj_uid SMALLINT UNSIGNED NOT NULL,\\\n predecessor SMALLINT UNSIGNED NOT NULL,\\\n successor SMALLINT UNSIGNED NOT NULL) ENGINE=MyISAM DEFAULT CHARSET=utf8;\nCREATE TABLE resources (proj_uid SMALLINT UNSIGNED NOT NULL,\\\n resource_uid SMALLINT UNSIGNED NOT NULL,\\\n calendar_uid SMALLINT UNSIGNED NOT NULL,\\\n resource_units DECIMAL(3,2) DEFAULT 1.00,\\\n resource_name VARCHAR(100),\\\n PRIMARY KEY (proj_uid, resource_uid)) ENGINE=MyISAM DEFAULT CHARSET=utf8;\nCREATE TABLE assignments (proj_uid SMALLINT UNSIGNED NOT NULL,\\\n task_uid SMALLINT UNSIGNED NOT NULL,\\\n resource_uid SMALLINT UNSIGNED NOT NULL,\\\n assignment_uid SMALLINT UNSIGNED NOT NULL,\\\n assignment_start DATETIME NOT NULL,\\\n assignment_finish DATETIME NOT NULL,\\\n assignment_work INT UNSIGNED DEFAULT 0,\\\n PRIMARY KEY (proj_uid, task_uid, resource_uid)) ENGINE=MyISAM DEFAULT CHARSET=utf8;\nCREATE TABLE assignment_data (proj_uid SMALLINT UNSIGNED NOT NULL,\\\n assignment_uid SMALLINT UNSIGNED NOT NULL,\\\n assignment_start DATETIME NOT NULL,\\\n assignment_finish DATETIME NOT NULL,\\\n assignment_work INT UNSIGNED DEFAULT 0) ENGINE=MyISAM DEFAULT CHARSET=utf8;\n<\/code><\/pre>\n\n\n\nThis SQL script sets up a database with several tables. The database can hold several project plans, but they all must have distinctive names. This SQL script only catches a subset of all information which is part of a Microsoft\u00ae<\/sup> Project<\/strong> plan in XML<\/strong> format; it does not capture calendar data, for example, nor do the data types in the SQL script necessarily match the ones from the specification in [1<\/a>]. But it is good enough to start with colored network graphs.<\/p>\n\n\n\n
This only needs to be executed one time. Once the database has been set up properly, we only have to execute Step 1 and Step 2.<\/p>\n\n\n\n
Step 1: Parsing the Project Plan<\/h3>\n\n\n\n
The first real step is to parse the Microsoft\u00ae<\/sup> Project<\/strong> plan in XML<\/strong> format; this is done with the bash script msp2mysql.sh<\/strong><\/a>. This program contains a finite-state machine<\/a> (see graph below), and the transitions between the states are XML tags (opening ones and closing ones). This approach helps to keep the code base small and easily maintainable. Furthermore, the code base can be enlarged if more XML tags or additional details of the project plan shall be considered.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/msp2mysql-456x1024.png\")
Finite-State Machine for parsing the Project Plan<\/figcaption><\/figure>\n\n\n\n I concede there are more elegant and faster ways to parse XML data, but I believe that msp2mysql.sh<\/a> can be understood well and helps to foster an understanding of what needs to be done in Step 1.<\/p>\n\n\n\n
There is one important limitation in the program that needs to be mentioned: The function iso2msp ()<\/strong> transforms a duration in ISO 8601 format to a duration described by multiples of 6 s (6 s are an old Microsoft\u00ae<\/sup> Project<\/strong> “unit” of a time duration, Microsoft\u00ae<\/sup> Project<\/strong> would traditionally only calculate with 0.1 min of resolution). However, as you can see in [2<\/a>], durations cannot only be measured in seconds, minutes, and hours, but also in days, months, etc. A duration of 1 month is a different amount of days, depending on the month we are looking at. This is something that has not been considered in this program; in fact, it can only transform durations given in hours, minutes and seconds to multiples of 6 s. Fortunately, this captures most durations used by Microsoft\u00ae<\/sup> Project<\/strong>, even durations that span over multiple days (which are mostly still given in hours in the Microsoft\u00ae<\/sup> Project<\/strong> XML file).<\/p>\n\n\n\n
While msp2mysql.sh<\/a> parses the Microsoft\u00ae<\/sup> Project<\/strong> plan in XML<\/strong> format, it typically runs through these steps:<\/p>\n\n\n\n
- It extracts the name of the project plan and checks if there is already a project plan with this name in the database. If that is the case, the program terminates. If that is not the case, the project name is stored in the table projects<\/strong>, and a new proj_uid<\/strong> is allocated.<\/li>
- It extracts details of the tasks and stored them in the tables tasks<\/strong> and task_links<\/strong>.<\/li>
- It extracts the resource names and stores their details in the table resources<\/strong>.<\/li>
- It extracts assignments (of resources to tasks) and stores them in the tables assignments<\/strong> and assignment_data<\/strong>.<\/li><\/ul>\n\n\n\n
msp2mysql.sh<\/a> is called with the Microsoft\u00ae<\/sup> Project<\/strong> plan in XML<\/strong> format as argument, like this:<\/p>\n\n\n\n
.\/msp2mysql.sh \/home\/tmp\/Example_1.xml<\/code><\/pre>\n\n\n\nStep 2: Generating a Graph<\/h3>\n\n\n\n
The second step is to extract meaningful data from the database and transform this into a colored network graph; this is done with the bash script mysql2dot.<\/strong><\/a>sh<\/a><\/strong>. mysql2dot.<\/strong><\/a>sh<\/a><\/strong> can create colored network graphs according to:<\/p>\n\n\n\n
- … the task duration<\/li>
- … the work allocated to a task<\/li>
- … the task slack<\/li>
- … the criticality of a task<\/li><\/ul>\n\n\n\n
mysql2dot.<\/a>sh<\/a> relies on a working installation of the graphviz<\/a> package, and in fact, it is graphviz<\/a> and its collection of tools that create the graph, while mysql2dot.<\/a>sh<\/a> creates a script for graphviz<\/a> in the dot<\/a><\/em> language. Let us look close to some aspects:<\/p>\n\n\n\n
mysql2dot.<\/a>sh<\/a> uses an uneven scale of colors, and the smallest scale should be three or more colors. The constant NUM_COLSTEPS sets the number of colors, in our case 7 colors. The constant COLSCHEME defines the graphviz<\/a> color scheme that shall be used, in our case this is “ylorrd7”. [3<\/a>] shows some really beautiful color schemes. The ColorBrewer schemes are well suited for our kind of purpose (or for physical maps); however, the license conditions ([4<\/a>]) must be observed.<\/p>\n\n\n\n
mysql2dot.<\/a>sh<\/a> is invoked with the name of a directory as argument. The generated script(s) in dot<\/a> language as well as the generated graph(s) will be stored in that directory. A typical call looks like this:<\/p>\n\n\n\n
.\/mysql2dot.sh \/home\/tmp\/<\/code><\/pre>\n\n\n\nWhen started, mysql2dot.<\/a>sh<\/a> retrieves a list of all projects from the database and displays them to the user who then has to select a project or choose the option Exit<\/em>. When a project plan has been selected, the user has to choose the graph that shall be generated. Subsequently, the program starts to generate a script in dot<\/a> language and invokes the program dot (a tool of the graphviz<\/a> suite) in order to compute the network graph.<\/p>\n\n\n\n
Different project plans typically have different ranges of durations, work or slack for all of their tasks. mysql2dot.<\/a>sh<\/a> therefore takes into account the maximum duration, the maximum work, or the maximum slack it can find in the whole project. Initially, I used a linear scale to distribute the colors over the whole range from [0; max_value], but after looking at some real project plans, I personally found that too many tasks would then be colored in darker colors and I thought that this would rather distract a project manager from focusing on the few tasks that really<\/em> need to be monitored closely. Trying out linear<\/em>, square<\/em> and square root<\/em> approaches, I finally decided for the square root<\/em> approach which uses the darker colors only when the slack is very small, the workload is very high, or the duration is very long. Consequently, the focus of the project manager is guided to the few tasks that really deserve special attention. However, this does not mean that this approach is the only and correct way to do it. Feel free to experience with different scaling methods to tune the outcome to your preferences.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/PERT-Farbskalen-1024x724.jpg\")
Different Algorithms and their Effect on Color Scales<\/figcaption><\/figure>\n\n\n\n The network graph colored according to task criticality<\/em> follows a different approach. Task criticality<\/em> is derived from the ration of task slack<\/em> and task duration<\/em>. Tasks which have a low ratio of task slack<\/em> \/ task duration<\/em> are defined as more critical than tasks where this ratio is higher. I also experienced with the square root approach mentioned above, but that seemed to result in different criticalities depending on the maximum ratio of task slack<\/em> \/ task duration<\/em> which seemed unsatisfactory to me. Finally, I opted for a logarithmic approach which puts the ration task slack<\/em> \/ task duration<\/em> = 1,0 to the middle value of NUM_COLSTEPS. So in the current case, a ratio of task slack<\/em> \/ task duration<\/em> = 1,0 (means: task slack<\/em> = task_duration<\/em>) would result in a value of 4.0. If NUM_COLSTEPS was 9 rather than 7, the result would be 5.0. This can be reached by the formula:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Formel-1.png\")
<\/figure>\n\n\n\nwhere:<\/p>\n\n\n\n
- n<\/em> is the number of tasks that are neither a milestone nor a summary task and that have a non-zero duration<\/li>
- j<\/em> is the task for which we determine the level of shade<\/li>
- number_of_shades<\/em> is the number of different shades you want to have in the graph; this should be an impair number and it should be 3 at least<\/li>
- shadej<\/sub><\/em> is the numerical value of the shade for task j<\/em><\/li>
- task_slackj<\/sub><\/em> is the task slack of task j<\/em><\/li>
- task_durationj<\/sub><\/em> is the duration of task j<\/em><\/li><\/ul>\n\n\n\n
Examples<\/h2>\n\n\n\n
Example 1: Sequential Task Chain<\/h3>\n\n\n\n
This example is a sequential task chain with 20 tasks; each task is +1d longer than the previous task.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-1024x256.jpg\")
Example 1: Sequential Task Chain<\/figcaption><\/figure>\n\n\n\n In this example, no resources have been assigned yet. As all tasks are in sequence, there is obviously no slack in any task, and all tasks are in the critical path. Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-1024x705.png\")
Parsing the project plan from Example 1<\/figcaption><\/figure>\n\n\n\n We can already see that the shall script takes a long time for this easy project plan. For a productive environment, one would have to use C code rather than a shell script, but for educational purposes, the shell script is easier to change, understand and also to debug. Subsequently, we start creating the graphs with mysql2dot.sh<\/strong>, here is an example on how this looks like:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Erzeugung-eines-Graphen-1024x587.png\")
Creation of a graph with mysql2dot.sh<\/figcaption><\/figure>\n\n\n\n Let us look at the 4 graphs which can be generated:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-duration-1024x17.png\")
Network Graph colored according to task duration (Example 1)<\/figcaption><\/figure>\n\n\n\n The result is what could be expected. Tasks with a shorter duration are colored in a lighter color, tasks with a longer duration consequently in a darker color. While the task duration increases in a linear fashion from left to right, the graph colors tasks with a longer duration more in the “dangerous” area, that is, in a darker color.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-work-1024x17.png\")
Network Graph colored according to task work (Example 1)<\/figcaption><\/figure>\n\n\n\n As no resource has been attributed, no work has been registered for any of the tasks. Therefore, all tasks are equally shown in a light color.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-slack-1024x17.png\")
Network Graph colored according to task slack (Example 1)<\/figcaption><\/figure>\n\n\n\n As none of the tasks has any slack, obviously all tasks are in the critical path. Hence all of them are colored in dark color.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_1-criticality-1024x17.png\")
Network Graph colored according to task criticality (Example 1)<\/figcaption><\/figure>\n\n\n\n Similar to the graph above, all tasks are critical, and progress must be carefully monitored; hence, all tasks are colored in dark color.<\/p>\n\n\n\n
Example 2: Parallel Task Chain<\/h3>\n\n\n\n
This example has the same tasks as Example 1, but all of them in parallel. The longest task therefore determines the overall project runtime.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-1024x724.jpg\")
Example 2: Parallel Task Chain<\/figcaption><\/figure>\n\n\n\n In this example, too, no resources have been assigned yet. Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-1024x705.png\")
Parsing the project plan from Example 2<\/figcaption><\/figure>\n\n\n\n The 4 graphs from this example look very different from Example 1:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-duration-128x1024.png\")
Network Graph colored according to task duration (Example 2)<\/figcaption><\/figure>\n\n\n\n The color of the tasks is the same as in Example 1; however, all tasks are in parallel.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-work-128x1024.png\")
Network Graph colored according to task work (Example 2)<\/figcaption><\/figure>\n\n\n\n The color of the tasks is the same as in Example 1 as we still have no work assigned to any of the tasks; however, all tasks are in parallel.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-slack-128x1024.png\")
Network Graph colored according to task slack (Example 2)<\/figcaption><\/figure>\n\n\n\n As to slack, the picture is different. Only one task (task #20) does not have any slack, all other tasks have more or less slack, and task #01 has the most slack.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_2-criticality-128x1024.png\")
Network Graph colored according to task criticality (Example 2)<\/figcaption><\/figure>\n\n\n\n Here, we can see that both tasks #19 and tasks #20 are considered to be critical. More tasks receive a darker color as compared to the graph considering slack alone.<\/p>\n\n\n\n
Example 3: Sequential Task Chain with Resource<\/h3>\n\n\n\n
This example has the same tasks as Example 1, but we assign a resource to each task (the same in this example, but this OK as all tasks are sequential, hence no overload of the resource). Consequently, work is assigned to each task, and we can expect changes in the network graph referring to task work.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-1024x256.jpg\")
Example 3: Sequential Task Chain with Resource<\/figcaption><\/figure>\n\n\n\n Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-1024x952.png\")
Parsing the project plan from Example 3<\/figcaption><\/figure>\n\n\n\n We can see that now (different from Example 1), resources and assignments are captured, too. Let us look at the 4 graphs from this example:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-duration-1024x15.png\")
Network Graph colored according to task duration (Example 3)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 1 as the durations are the same.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-work-1024x15.png\")
Network Graph colored according to task work (Example 3)<\/figcaption><\/figure>\n\n\n\n This graph now experiences the same coloring as the network graph according to task duration. That is understandable given the fact that we have a full-time resource working 100% of the time for the full task duration, hence work and duration have the same numeric values for each respective task.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-slack-1024x15.png\")
Network Graph colored according to task slack (Example 3)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 1 as still none of the tasks has any slack.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_3-criticality-1024x15.png\")
Network Graph colored according to task criticality (Example 3)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 1 as still none of the tasks has any slack. Consequently, all of them are critical.<\/p>\n\n\n\n
Example 4: Parallel Task Chain with Resources<\/h3>\n\n\n\n
This example has the same tasks as Example 2, but we assign a resource to each task. As all tasks are parallel, we assign a new resource to each task so that there is no resource overload.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-1024x724.jpg\")
Example 4: Parallel Task Chain with Resources<\/figcaption><\/figure>\n\n\n\n Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-951x1024.png\")
Parsing the project plan from Example 4<\/figcaption><\/figure>\n\n\n\n We can see that now (different from Examples 2), resources and assignments are captured, too. Let us look at the 4 graphs from this example:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-duration-139x1024.png\")
Network Graph colored according to task duration (Example 4)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 2 as the durations are the same.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-work-139x1024.png\")
Network Graph colored according to task work (Example 4)<\/figcaption><\/figure>\n\n\n\n This graph now experiences the same coloring as the network graph according to task duration. That is understandable given the fact that we have a full-time resource working 100% of the time for the full task duration, hence work and duration have the same numeric values for each respective task.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-slack-139x1024.png\")
Network Graph colored according to task slack (Example 4)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 3; the slack of each task has remained the same.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_4-criticality-139x1024.png\")
Network Graph colored according to task criticality (Example 4)<\/figcaption><\/figure>\n\n\n\n Nothing has changed compared to Example 3; the criticality of the tasks remains the same.<\/p>\n\n\n\n
Example 5: Sequential Blocks with Parallel Tasks and with Resources<\/h3>\n\n\n\n
This example is the most interesting one as it combines different aspects within one project plan. We have 5 milestones, and 4 parallel tasks before each milestone. All tasks have resources assigned.
Task Block 1<\/strong> contains 4 tasks of different durations, and as each task has a 100% resource allocated, also of different work. Due to the different durations, the 4 tasks also have different slack, with the longest task having zero slack.
Task Block 2<\/strong> is like Task Block 1, but the work is the same for each task; this has been achieved by an over-allocation of the respective resources for 3 out of the 4 tasks.
Task Block 3<\/strong> consists of 4 parallel tasks with the same duration, however with a work allocation similar to Task Block 1. None of the tasks therefore has slack, but the work for each task is different.
Task Block <\/strong>4 is like Task Block 1 in terms of task duration and task work. However, the tasks have been arranged in a way so that none of the tasks has slack.
Task Block 5<\/strong> is like Task Block 4, but each task has (the same amount of) lag time between the task end and the subsequent milestone. None of the tasks has any slack.<\/p>\n\n\n\n![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-1024x724.jpg\")
Example 5: Sequential Blocks with Parallel Tasks and with Resources<\/figcaption><\/figure>\n\n\n\n Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-951x1024.png\")
Parsing the project plan from Example 5<\/figcaption><\/figure>\n\n\n\n Let us look at the 4 graphs from this example:<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-duration-1024x245.png\")
Network Graph colored according to task duration (Example 5)<\/figcaption><\/figure>\n\n\n\n The colors in Task Block 1, 2, 4 and 5 are the same. Only in Task Block 3 where all tasks have the same (long) duration, all tasks have a dark color.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-work-1024x245.png\")
Network Graph colored according to task work (Example 5)<\/figcaption><\/figure>\n\n\n\n The colors in Task Block 1, 3, 4, and 5 are the same. Only in Task Block 2 where all tasks have the same amount of work (which is also the highest work a task has in this project plan), all tasks have a dark color.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-slack-1024x245.png\")
Network Graph colored according to task slack (Example 5)<\/figcaption><\/figure>\n\n\n\n The colors in Task Block 1 and 2 are the same. The tasks in Task Blocks 3, 4, and 5 are all dark as none of these tasks has any slack.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_5-criticality-1024x245.png\")
Network Graph colored according to task criticality (Example 5)<\/figcaption><\/figure>\n\n\n\n The network graph according to task criticality is similar to the one according to slack, but we can see that in Task Blocks 1 and 2 tasks get a darker color “earlier” as in the network graph according to criticality.<\/p>\n\n\n\n
Example 6: Distinguishing Slack and Criticality<\/h3>\n\n\n\n
So far, we have not seen much difference between the network graphs according to slack and the one according to criticality. In this example, we have two task blocks with 4 tasks each where the slack is the same in the corresponding tasks in each task block. However, the duration is different between both task blocks. Work per task is almost the same, I tried to match the work despite the longer duration of the tasks in Task Block 2 as good as possible.<\/p>\n\n\n\n
![\"\"](\"https:\/\/caipirinha.spdns.org\/wp\/wp-content\/uploads\/Example_6-1024x378.jpg\")
Example 6: Distinguishing Slack and Criticality<\/figcaption><\/figure>\n\n\n\n Parsing this project plan with msp2mysql.sh<\/strong> leads to:<\/p>\n\n\n\n
- j<\/em> is the task for which we determine the level of shade<\/li>
- n<\/em> is the number of tasks that are neither a milestone nor a summary task and that have a non-zero duration<\/li>
- It extracts details of the tasks and stored them in the tables tasks<\/strong> and task_links<\/strong>.<\/li>
- … have the package graphviz<\/a> installed<\/li>