Delay probability prediction method for key chain plan nodes

Abstract

The invention discloses a delay probability prediction method for a key chain plan node, and the method is characterized in that the method comprises the following steps: S1, carrying out the plan execution state data collection; s2, node execution state data conversion is carried out, and plan execution state data is converted into structured graph data; s3, a project progress network diagram is constructed, and the project progress network diagram is constructed according to the tight-front and tight-back relation between the nodes; s4, project real-time key chain identification is carried out, the remaining construction period of each plan chain is counted by using a graph search algorithm, and the link with the longest remaining construction period is a key chain; and S5, key chain plan node delay probability prediction is carried out, the key chain plan node delay probability is calculated based on the execution state of the nodes on the key chain, and the demand of plan execution state dynamic perception is met. According to the method, the delay probability of the key chain plan node is monitored in real time in the plan execution process, and the method has important significance in improving the project plan execution efficiency and avoiding project delay.

Description

Technical Field

[0001] The present invention relates to the technical field of project management, and specifically to a method for predicting the delay probability of critical chain planning nodes. Background Art

[0002] Project schedule management refers to managing a series of processes required for a project to ensure its timely completion. Project schedule status perception is an important management process in the schedule control link, which includes constructing plan warning rules based on multi-level project plan information, establishing the correlation between multi-dimensional data such as actual completion time, planned completion time, and node relationships, and predicting the plan delay situation according to the current project time node and schedule status. The purpose is to point out the direction for avoiding project risks and improving the efficiency of project plan management.

[0003] The critical chain technique helps to accurately perceive the project schedule status. By identifying the task sequences that directly affect the project completion date in project schedule plan management, these tasks are defined as critical chain planning nodes and controlled. Since the deviations on the critical chain will directly affect the project duration, evaluating the progress of critical chain activities helps to identify project delay risks.

[0004] The satellite AIT process contains a large number of experimental steps, there are resource dependencies between the tests, and there are a large number of precedence relationships between the processes. Therefore, the critical chain technique is applicable to the satellite AIT process plan. In the past, the satellite AIT project schedule status could only be presented in the form of tables or statistics, which made it impossible for project schedulers to intuitively identify the delay risks of critical chain planning nodes and the overall project delay risks, thereby affecting the on-time completion rate of the plan.

[0005] After retrieval, 3 relatively close prior art documents were found:

[0006] (1) Patent document: A project schedule management method based on the critical chain (application number: CN201410024924.5). In this document, the critical chain (Critical Chain Project Management, CCPM) method and the schedule arrangement model under multi-resource constraints are used to formulate the project plan, and the project is monitored by the buffer consumption percentage.

[0007] In contrast, in the present invention, the critical chain is identified in real time during the execution of the project plan, and then methods such as tolerance chains are used to calculate the delay probabilities of each planning node on the critical chain, so as to achieve the purpose of quantitatively monitoring the delay risks of the planning nodes. It is not used for formulating the project plan, and there are differences in the way of monitoring the project execution situation.

[0008] (2) Patent Literature: A project management method based on the critical chain of TOC (Application No.: CN202010309896.7). The method proposed in this literature is mainly used to obtain the optimal construction path planning, preset buffer time, and reduce the overdue risk.

[0009] In contrast, the ultimate goal of the present invention is to quantitatively predict the delay probability of planned nodes. There are significant differences between this literature and the purpose and method of the present invention.

[0010] (3) Journal Literature: Li Cunbin, Xu Liang. Research on the transfer of project duration risk based on the critical chain [J]. Project Management Technology, 2013, 2: 33-38. The purpose of this literature is to identify bottleneck activities by using the law of risk transfer and improve the size of the buffer.

[0011] In contrast, the present invention aims to achieve the prediction of the delay probability of critical chain planned nodes, rather than calculating the expected delay and identifying bottleneck activities. Moreover, there are differences in the calculation methods of risk transfer between the present invention and this literature. The present invention uses tolerance chain and probability distribution methods to simulate risk transfer, while this literature uses risk absorption coefficient to calculate the risk transfer of each process on the chain. Summary of the Invention

[0012] The purpose of the present invention is to provide a method for predicting the delay probability of critical chain planned nodes, which is used to monitor the delay probability of critical chain planned nodes in real time during the plan execution process and continuously and dynamically perceive the project execution status.

[0013] To achieve the above purpose, the present invention provides a method for predicting the delay probability of critical chain planned nodes, which is characterized by including the following steps:

[0014] S1: Collect data on the plan execution status and obtain the execution status data of each plan node;

[0015] S2: Convert the node execution status data and transform the plan execution status data into structured graph data;

[0016] S3: Construct a project schedule network diagram and construct a project schedule network diagram according to the precedence relationship between nodes;

[0017] S4: Identify the real-time critical chain of the project, use the graph search algorithm to identify the critical chain in the network diagram in real time, that is, use the graph search algorithm to calculate the remaining duration of each planned chain, and the link with the longest remaining duration is the critical chain;

[0018] S5: Predict the delay probability of critical chain planned nodes, calculate the delay probability of critical chain planned nodes based on the execution status of nodes on the critical chain, and meet the requirements of dynamically perceiving the plan execution status.

[0019] Furthermore, the planned execution status data in step S1 includes the outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion degree of each planned node.

[0020] Furthermore, in step S2, the conversion of node execution status data is to store each planned node as a node in the graph database, and use the outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion degree as the basic attributes of each node.

[0021] Furthermore, step S3 includes: initially determining node relationships through the outline level number, and forming a task sequence in ascending order of serial numbers for nodes under the same section outline; supplementing and determining node relationships through resource and time information. If the day after the planned completion time of node A is the planned start time of node B, and there is a resource dependency between node A and node B, then node B should be regarded as the immediate successor activity of node A; adding a start node and an end node, pointing the start node to all starting points of the planned chains, and pointing all ending points of the planned chains to the end node.

[0022] Furthermore, step S4 includes: using a graph search algorithm to identify all paths in the project plan; taking the current time as the baseline, identifying the completed nodes and uncompleted nodes, and calculating the remaining duration of the link (link remaining duration calculation); identifying the path with the longest remaining duration as the real-time critical chain.

[0023] Furthermore, the link remaining duration calculation takes the current time as the baseline, calculates the remaining duration of each link according to the real-time completion degree of the nodes, and identifies the link with the longest remaining duration as the real-time critical chain. The calculation formula is:

[0024] Link remaining duration = ∑(duration of uncompleted nodes) × (1 - completion degree).

[0025] Furthermore, in step S5, tolerance chain, unified compression, and normal distribution techniques are used to calculate the delay probability of the critical chain planned nodes.

[0026] Furthermore, step S5 includes: using the unified compression method to allocate 20% of the planned duration of each node for buffer time, unifying the standard working hours to 80% of the planned duration, and setting the earliest completion time of the nodes that are being delayed and those that are in progress and not delayed; using the tolerance chain method to calculate the earliest start time and earliest completion time of each node; calculating the delay probability of the critical chain planned nodes according to the normal distribution.

[0027] Furthermore, the formula for calculating the delay probability of the nodes in step S5 is:

[0028]

[0029] Furthermore, in step S5, the difference between the earliest start time later than the planned start time is used as the basis for calculating the delay probability. The greater the ratio of this difference to the planned duration, the greater the node delay probability. The range of this ratio is converted into the range of a random variable that follows a standard normal distribution to calculate the delay probability.

[0030] The present invention helps project schedulers to monitor the delay probability of critical chain plan nodes in real time during the plan execution process, dynamically perceive the change of plan execution status, continuously master the critical chain warning information, and has important significance for improving the project plan execution efficiency and avoiding project delays.

[0031] Compared with the prior art, the present invention has the following advantages: The present invention realizes a method for predicting the delay probability of critical chain plan nodes. Compared with the traditional project execution status monitoring methods such as plan completion rate statistics and buffer consumption percentage statistics, by predicting the delay probability of plan nodes on the critical chain in real time, it can quantitatively identify project delay risks, help project schedulers dynamically perceive the change of plan execution status, continuously master the critical chain warning information, and allocate inputs targeted according to the delay risks of nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flowchart of the method for predicting the delay probability of critical chain plan nodes of the present invention.

[0033] Figure 2 It is a schematic diagram of determining node relationships using outline levels in an example of the present invention.

[0034] Figure 3 It is a schematic diagram of determining node relationships using resource time information in an example of the present invention.

[0035] Figure 4 It is a progress network diagram of the whole satellite AIT project in an example of the present invention.

[0036] Figure 5 It is a diagram of the automatic identification result of the critical chain in an example of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] The following further describes the specific embodiments of the present invention in detail with reference to the accompanying drawings.

[0038] To make the objectives, technical solutions and advantages of the present invention clearer, the following will be combined with Figures 1 to 5A further detailed description is given to the method for predicting the delay probability of the critical chain plan nodes of the present invention. It should be noted that many details listed in the specification are only for enabling the reader to have a more thorough understanding of one or more aspects of the present invention, and these methods of the present invention can be implemented even without these specific details.

[0039] The method for predicting the delay probability of the critical chain plan nodes of the present invention converts the plan execution status data into structured graph data, constructs a project schedule network diagram according to the precedence relationship between nodes, uses a graph search algorithm to identify the critical chain in the network diagram in real time, and calculates the delay probability of the critical chain plan nodes based on the execution status of the nodes on the critical chain, meeting the requirement of dynamically perceiving the plan execution status.

[0040] Figure 1 This is a flowchart of the method for predicting the delay probability of the critical chain plan nodes of the present invention. As Figure 1 shown, the method for predicting the delay probability of the critical chain plan nodes of the present invention has the following steps:

[0041] Step 1: Collection of plan execution status data.

[0042] Obtain the completion status attributes of the plan nodes.

[0043] The plan execution status data includes the plan name, planned duration, outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion degree of each plan node. The project execution personnel regularly update and maintain the above information after the plan starts.

[0044] Step 2: Conversion of node execution status data.

[0045] Store each plan node as a node in the graph database, and store the execution status collected in Step 1 as the attributes of the node.

[0046] Step 3: Construction of the project schedule network diagram.

[0047] Infer the precedence relationship between nodes using information such as names and times.

[0048] The present invention determines the precedence relationship between nodes according to the outline number, resource, and time information to construct a project schedule network diagram. Determining the precedence relationship between nodes according to the outline number means that nodes belonging to the same section outline can form a task sequence in ascending order of serial numbers. Determining the precedence relationship between nodes according to the resource and time information means that there is a resource dependency between two plan nodes, and if the planned completion time of one node is adjacent to the planned start time of the other node, there should be a precedence relationship between the two nodes.

[0049] Step 4: Identification of the real-time critical chain of the project.

[0050] Use the graph search algorithm to calculate the remaining duration of each planned chain. The chain with the longest remaining duration is the critical chain.

[0051] The calculation of the remaining duration of the chain is based on the current time as the baseline, and calculates the remaining duration of each chain according to the real-time completion degree of the nodes. The calculation formula is:

[0052] Remaining duration of the chain = ∑(Duration of uncompleted nodes) × (1 - Completion degree)

[0053] For example, the real-time critical chain identification process of the project is as follows: Use the graph search algorithm to identify all paths in the project plan; Based on the current time as the baseline, identify the completed nodes and uncompleted nodes, and calculate the remaining duration of the chain; Identify the path with the longest remaining duration as the real-time critical chain.

[0054] By using the graph search algorithm for depth-first traversal, the path set of all project planned chains can also be obtained, and the total duration of each chain is obtained by accumulating the durations of each path in the set.

[0055] Step 5: Prediction of the delay probability of critical chain planned nodes.

[0056] Use tolerance chain, unified compression, and normal distribution techniques to calculate the delay probability of critical chain planned nodes.

[0057] As a further technical solution of the present invention: The calculation process of the delay probability of the critical chain planned node is as follows:

[0058] Use the unified compression method to allocate 20% of the planned duration of each node for buffer time, unify the standard working hours to 80% of the planned duration, and set the earliest completion time of the nodes that are being delayed and the nodes that are in progress and not delayed; Use the tolerance chain method to calculate the earliest start time and earliest completion time of each node; Calculate the delay probability of critical chain planned nodes according to the normal distribution.

[0059] The calculation formula for the delay probability of the node is as follows:

[0060]

[0061] Example

[0062] Use the delay probability prediction method for critical chain planned nodes of the present invention for experimental verification.

[0063] Step 1: Collection of planned execution status data

[0064] The planned execution status data of this example is sourced from the regular reporting of a certain satellite AIT process plan by the planned execution personnel. This data is stored in a database in list form, and the field information to be collected includes outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion percentage.

[0065] Step 2: Node Execution Status Data Conversion

[0066] Sort out the node information from the planned execution status data and construct the node execution status. Extract the planned nodes with the finest outline level in the planned data as independent objects for storage, and use the six fields of node name, planned start time, planned completion time, actual start time, actual completion time, and completion percentage as the basic attributes of each node.

[0067] Step 3: Construction of Project Schedule Network Diagram

[0068] Construct a project schedule network diagram by identifying whether there is a precedence and successor relationship between two nodes in the satellite AIT process. This example is determined through the following three steps:

[0069] (1) Initially determine the node relationship through the outline level number. Nodes belonging to the same section outline can form a task sequence in ascending order of serial number. The relationships that can be determined by the outline level in a certain satellite AIT plan are as Figure 2 shown.

[0070] (2) Supplement and determine the node relationship through resource and time information. If the day after the planned completion time of node A is the planned start time of node B, and there is a resource dependency between node A and node B, then node B should be regarded as the successor activity of node A. The relationships that can be determined by time supplement in a certain satellite AIT plan are as Figure 3 shown.

[0071] (3) Add a start node and an end node. Point the start node to the starting point of all planned chains, and point the ending point of all planned chains to the end node. The node relationship of a certain satellite AIT plan is as Figure 4 shown.

[0072] Step 4: Identification of Project Real-Time Critical Chain

[0073] Continuously update the project critical chain as the plan progresses, dynamically master the critical chain information, and determine it through the following three steps:

[0074] (1) Project Plan Chain Identification:

[0075] The project plan chain is a path that consists of a series of nodes with a precedence and successor relationship from the start node to the end node. The present invention uses a graph search algorithm for depth traversal to obtain the path set of all project plan chains, and accumulates the construction periods for each path in the set to obtain the total construction period of each link.

[0076] (2) Calculation of the remaining construction period of the link:

[0077] Taking the current time as the baseline, identify the completed nodes and uncompleted nodes. In order to monitor the completion status of each plan chain in real time, for each link, use the following formula to calculate the remaining construction period of the link.

[0078] Remaining construction period of the link = ∑(Construction period of uncompleted nodes) × (1 - Degree of completion)

[0079] (3) Identification of the critical chain:

[0080] Identify the link with the longest remaining construction period as the real-time critical chain. The automatically identified critical chain result is as Figure 5 shown, and the construction period of the critical chain can be automatically calculated to be 151 days.

[0081] Step 5: Prediction of the delay probability of the critical chain plan nodes

[0082] Quantitatively identify the project delay risk by predicting the delay probability of the critical chain plan nodes, help project schedulers dynamically perceive the changes in the plan execution status, continuously master the critical chain warning information, and allocate inputs targeted according to the delay risk of the nodes. It is determined through the following three steps:

[0083] (1) Use the unified compression method to allocate 20% of the planned construction period of each node for buffer time, and unify the standard working hours to 80% of the planned construction period. Define the earliest completion time of all nodes that are currently delayed as the current time, and define the earliest completion time of all nodes that are in progress and not delayed as the actual start time plus the standard working hours;

[0084] (2) Use the tolerance chain method to calculate the earliest start time and earliest completion time of each node: The earliest start time is equal to the maximum value of the earliest completion times of all the predecessor activities of the node, and the earliest completion time is equal to the earliest start time plus the standard working hours of the node. By analogy, for each plan chain, calculate the earliest start time and earliest completion time of each node in order from front to back according to the node sequence;

[0085] (3) In this example, the difference between the earliest start time and the planned start time is used as an important basis for calculating the delay probability. Considering that the durations of different planned nodes are different, the greater the ratio of this difference to the planned duration, the greater the delay probability of the node. The range of this ratio is converted into the range of a random variable subject to the standard normal distribution to calculate the delay probability. The specific method is as follows: Determine whether the earliest start time of the node is later than the planned start time. If it is later than the planned start time, the formula for calculating the delay probability of this node is as follows:

[0086]

[0087] The prediction results of the delay probabilities of the critical chain plan nodes on the 100th day of the plan execution of the present invention are shown in Table 1.

[0088] Table 1 Prediction Results of Delay Probabilities of Critical Chain Plan Nodes

[0089]

[0090]

[0091] It should be noted that the above is only a schematic description and elaboration of the present invention. Those skilled in the art should understand that any modification and replacement of the present invention fall within the protection scope of the present invention.

Claims

1. A method for predicting the probability of delay for critical chain planning nodes, characterized in that, It includes the following steps: S1: Collect the planned execution status data to obtain the execution status data of each planned node; S2: Convert the node execution status data to convert the planned execution status data into structured graph data; S3: Construct a project schedule network diagram and construct a project schedule network diagram according to the precedence relationship between nodes; S4: Identify the real-time critical chain of the project, use the graph search algorithm to identify the critical chain in the network diagram in real time, that is, use the graph search algorithm to calculate the remaining duration of each planned chain, and the link with the longest remaining duration is the critical chain; S5: Predict the delay probability of the critical chain planned node, calculate the delay probability of the critical chain planned node based on the execution status of the nodes on the critical chain, and meet the requirements of dynamic perception of the planned execution status.

2. The method for predicting the delay probability of a key-chain planning node according to claim 1, wherein The planned execution status data in step S1 includes the outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion degree of each planned node.

3. The method for predicting the delay probability of a key-chain plan node according to claim 1, characterized in that In step S2, the node execution status data conversion is to store each planned node as a node in the graph database, and use the outline level, node name, planned start time, planned completion time, actual start time, actual completion time, and completion degree as the basic attributes of each node.

4. The method for predicting the delay probability of the critical chain planning node according to claim 1, characterized in that Step S3 includes: initially determining the node relationship by the outline level number, and forming a task sequence for the nodes under the same section outline in ascending order of serial number; supplementing and determining the node relationship through resource and time information. If the day after the planned completion time of node A is the planned start time of node B, and there is a resource dependency between node A and node B, then node B should be regarded as the successor activity of node A; add a start node and an end node, point the start node to the starting point of all planned chains, and point the end points of all planned chains to the end node.

5. The method for predicting the delay probability of a key-chain planning node according to claim 1, wherein Step S4 includes: using the graph search algorithm to identify all paths in the project plan; taking the current time as the baseline, identifying the completed nodes and uncompleted nodes, and calculating the remaining duration of the link; identifying the path with the longest remaining duration as the real-time critical chain.

6. The method for predicting the delay probability of a critical chain planning node according to claim 5, characterized in that The calculation of the remaining duration of the link is based on the current time as the baseline, calculates the remaining duration of each link according to the real-time completion degree of the node, and identifies the link with the longest remaining duration as the real-time critical chain. The calculation formula is: Remaining duration of the link = ∑(Duration of uncompleted nodes) × (1 - Completion degree).

7. The method for predicting the delay probability of a key chain planning node according to claim 1, wherein In step S5, the tolerance chain, unified compression, and normal distribution techniques are used to calculate the delay probability of the critical chain planned node.

8. The method for predicting the delay probability of the critical chain planning node according to claim 7, characterized in that Step S5 includes: using the unified compression method to use 20% of the planned duration of each node for buffer time, unifying the standard working hours to 80% of the planned duration, and setting the earliest completion time of the nodes that are being delayed and are in progress and not delayed; using the tolerance chain method to calculate the earliest start time and earliest completion time of each node; calculating the delay probability of the critical chain planned node according to the normal distribution.

9. The method for predicting the delay probability of a key-chain planning node according to claim 1, characterized in that The calculation formula for the delay probability of the node in step S5 is:

10. The method for predicting the delay probability of a key chain planning node according to claim 1, characterized in that, In the step S5, the difference between the earliest start time later than the planned start time is used as the basis for calculating the delay probability. The larger the ratio of this difference to the planned construction period, the greater the node delay probability. The range of this ratio is converted into the range of a random variable subject to the standard normal distribution to calculate the delay probability.

Images