By Mingdao Cloud Sales Consultant Xiong Min

I have come into contact with many specific scenarios of different customers. When applying the basic functions of Mingdao Cloud to combine and realize these different scenarios, I will suddenly find that they are all in the same relationship. When using function points, the configuration logic is basically the same. Next, we will share the specific customer scenario of “automatic allocation of payment order amount according to contract node”.

1. Scene requirement analysis

Contract management with multiple nodes, each node has a corresponding contract for gathering data, customer payment after receipt of the invoice will be on, financial do receivable registration and the need to increase the amount of the collection of the contract amount and the corresponding node (the date of the contract node sequence) in sequence matching, deducted operations, until you assign the amount of the receivable.

When I learned about the scene over the phone, I was actually quite confused about the relationship between data objects, so I communicated again. While communicating, I wrote data docking with paper and pen, and helped to understand the corresponding relationship diagram, and initially opened the data objects.

2. Initial landing of demand

Comb after completing data relations, found in this scenario, it is difficult in the process design, need the initial data (the receivable amount) and standard data node (contract) (the date of the contract node sequence) in sequence comparison, deduction, each round after the deduction of the rest of the data and data contrast, next again this cycle.

After understanding this, it was almost the same as the performance rating interval allocation service we did for a customer with performance management needs a few days ago, and even a little simpler. Therefore, I consulted our implementation student Jiang Hao and began to implement the demand through the system.

Key functions: sequential execution of sub-processes and use of process parameters.

System configuration: Landing scheme

1) Worksheet configuration:

First, complete the configuration of the data store table [worksheet]. It is very simple. Configure the association relationship of the four data objects.

2) Workflow configuration:

Validation: Enter simple data to execute process validation results

Enter simple data to verify the flow configuration.

Identify problems/obstacles

Input data to perform the test, the error is found in the data results significantly and theory, through understanding of the system function, combining view workflow execution history, working table log to find problems, the reason is the key function is not as the first process parameter 】 【 comparison operation and updated with the latest value, used by the next comparison operations, It is always the original data (the amount of payment collected) that enters the process calculation.

Landing encountered obstacles, thinking for key objects

The remaining amount of each comparison operation cannot be used by [process parameters], so how can such data be used? The solution is to write the calculation results into the form, enter the flow for use, and update the data once a time.

The configuration is as follows:

1) “Payment Collection Worksheet” : added a field [subprocess reference calculated value];

2) workflow: process parameters are used to find data, not store data; Update node adds updates to the secondary subprocess reference calculation field.

The logical configuration of the two schemes is the same.

1) Worksheet comparison: added replication field of stored data [subprocess reference calculation]

2) Workflow comparison:

4, inspired

This demand landing process and has been “disassembly process, add intermediate data, to simplify” consistent with the idea.

For example, we used this idea to solve the periodic trigger reminder and summary process on the last day of each month. Difficult points:

1) Objective fact: The date of the last day of each month is not fixed.

2) Tool limitation: the timing trigger of the current system does not support date calculation.

Step 1: Solve the first problem: how to calculate the last day of the month;

Step 2: Think about how you can apply what you already have. We know the calculation method, but the timing trigger does not support calculation, so we can directly calculate the data of the last day, how to use it;

Step 3: Disassemble. The system has the function of triggering by date field, which can write the calculation data into the form, which can be used as the trigger and condition of “timing trigger”.