Developing

LICENSE

@@ -186,7 +186,7 @@
       same "printed page" as the copyright notice for easier
       identification within third-party archives.
 
-   Copyright [yyyy] [name of copyright owner]
+   Copyright [2023] [Longxing Tan]
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

Makefile

@@ -1,4 +1,3 @@
-
 .PHONY: style test docs
 
 check_dirs := lekin examples tests
@@ -8,8 +7,8 @@ check_dirs := lekin examples tests
 style:
 	black --preview $(check_dirs)
 	isort $(check_dirs)
-	flake8
-	pre-commit run --all-files
+	flake8 $(check_dirs)
+	pre-commit run --files $(check_dirs)
 
 # run tests for the library
 

README.md

@@ -34,9 +34,22 @@
 
 - accelerate by
 - Changeover Optimization
-- Ready for demo, research and production
+- Ready for demo, research and maybe production
 
 # DRATFING- NOT FINISHED AND DON'T USE IT NOW!
+
+## Feature
+
+- constrained optimization
+  - route
+  - production
+  - material kit
+  - together
+
+- soft constrained optimization
+  - objective
+
+
 ## Tutorial
 
 **Installation**
@@ -67,6 +80,7 @@ In real world, Lekin integrates with MES to deploy production plans on the shop
 
 - Construction heuristics
     - [SPT]()
+    - [critical path]()
 
 - Meta heuristics
     - [local search]()

README_zh_CN.md

@@ -27,15 +27,22 @@
 
 **[文档](https://python-lekin.readthedocs.io)** | **[教程](https://python-lekin.readthedocs.io/en/latest/tutorials.html)** | **[发布日志](https://python-lekin.readthedocs.io/en/latest/CHANGELOG.html)** | **[English](https://github.com/LongxingTan/python-lekin/blob/master/README.md)**
 
-**python-lekin**是一个数字化车间智能排产调度工具，名字来源于[教育工具Lekin](https://web-static.stern.nyu.edu/om/software/lekin/)
+**python-lekin**是一个APS智能排产调度工具，名字来源于[Lekin](https://web-static.stern.nyu.edu/om/software/lekin/)。在考虑实际约束的前提下，动态调整计划排程，高效响应客户订单承诺。
+
+## 特征
+
+- 支持工艺路线约束
+- 支持产能约束
+- 支持物料齐套约束
+- 支持顺排、倒排等方法
 
 
 ## 快速入门
 
 ### 安装
 
 ``` shell
-$ pip install python-lekin
+$ pip install lekin
 ```
 
 ### 使用
@@ -57,8 +64,13 @@ scheduler.draw()
 - Route
 
 ## 示例
+在实际APS系统开发中，
+
+- 按工艺路线拆分工序
+- 按BOM拆分物料
 
 ## 引用
+
 ```
 @misc{python-lekin2022,
   author = {Longxing Tan},

docs/source/api.rst

@@ -10,4 +10,4 @@ API
 
    scheduler
    solver
-   struct
+   lekin_struct

docs/source/application.rst

@@ -1,7 +1,7 @@
 Application
 ===========
 
-Data Requirement
+数据
 ----------------
 
 MRP: Material Requirements Planning
@@ -10,9 +10,26 @@ MRP: Material Requirements Planning
 BOM: Bill Of Materials
 
 
-Function requirement
+功能
 ------------------------
 
+一个完善的APS系统包含以下模块。
 - 需求中心
 - 排程中心
 - 排程工作台
+- 物料中心
+
+
+理解
+----------
+
+实际APS会涉及到各个车间，各个工序的复杂区分，以及BOM中涉及到多工厂的部分。
+- 其实和APS关系不大，APS只要把最后一层工序展开，最后一层BOM展开，按实际的资源约束进行计算，最后只是展现形式上的区别
+
+首先解决m个工序、n个机器的车间排产问题，然后把实际问题往车间问题靠。
+
+
+可视化
+------------
+- 资源在时间线上的计划情况
+- 按订单，在时间线上的操作情况

docs/source/constraints.rst

@@ -9,5 +9,3 @@ Constraints
 subject to:
 
 .. math:: \alpha{}_t(i) = P(O_1, O_2, … O_t, q_t = S_i \lambda{})
-
-

docs/source/index.rst

@@ -10,7 +10,13 @@ python-lekin documentation
    <a class="github-button" href="https://github.com/LongxingTan/python-lekin" data-icon="octicon-star" data-size="large" data-show-count="true" aria-label="Star LongxingTan/python-lekin on GitHub">GitHub</a>
 
 
-**python-lekin** is a Flexible Job shop Scheduler Library, named after and inspired by `Lekin <https://web-static.stern.nyu.edu/om/software/lekin/>`_.
+**python-lekin** 是一个工厂智能排产调度工具，名字来源于`Lekin <https://web-static.stern.nyu.edu/om/software/lekin/>`_.
+
+快速入门
+---------
+
+排产是一个分配任务，将有限的资源分配给需求。因此需求需要有优先级，约束主要有产能约束与物料约束。产能约束，将订单中的成品按工艺路线分解为工序，而每一道工序有对应的生产机器；物料约束，将订单的成品按BOM(bill of materials)展开为原材料需求，每一道工序开始前需要对应原材料齐套。
+
 
 Finite Capacity Planning
 
@@ -20,6 +26,7 @@ Finite Capacity Planning
 
    tutorials
    models
+   constraints
    application
    api
    CHANGELOG

docs/source/models.rst

@@ -19,3 +19,42 @@ SPT—EDD规则
 2）计算所有任务的总流程时间。
 3）查找方案中，预定交贷期（di）大于总流程时间的生产任务（不惟一），按SPT规则，将其中加工时间最大者排于最后。
 4）舍弃第3步能排定的最后任务者及其后序任务，回到第2步重复。
+
+
+关键路径法
+-------------
+
+关键路径是决定项目完成的最短时间，关键路径可能不止一条。
+
+其基本概念：
+最早开始时间 (Early start)
+最晚开始时间 (Late start)
+最早完成时间 (Early finish)
+最晚完成时间 (Late finish)
+松弛时间 (slack)
+
+正推方法确定每个任务的最早开始时间和最早完成时间，逆推方法确定每个任务的最晚完成时间和最晚开始时间。
+
+
+倒排
+---------------
+
+- 每一个MO最早开始时间初始化：max(ESD, today)。确保开始时间不早于今天
+
+
+正排
+-------------
+
+
+
+
+禁忌搜索
+------------
+
+
+遗传算法
+-------------
+
+
+强化学习
+-------------

lekin/__init__.py

@@ -5,7 +5,7 @@
 # from lekin.lekin_struct.operation import Operation
 # from lekin.lekin_struct.route import Route
 from lekin.scheduler import Scheduler
-from lekin.solver.dispatching_rules import Rules
+from lekin.solver.construction_heuristics import Rules
 from lekin.solver.meta_heuristics import Heuristics
 
 __all__ = ["Job", "Machine", "Route", "Operation", "Scheduler", "get_data"]

lekin/dashboard/pages.py

@@ -0,0 +1 @@
+import streamlit as st

lekin/lekin_struct/__init__.py

@@ -1,8 +1,15 @@
 """
 https://gitee.com/EnCode/APS?_from=gitee_search#%E5%BB%BA%E6%A8%A1
+
+工厂：订单
+车间：工单
+班组：工序
+操作动作：动作
+
 """
 
 from lekin.lekin_struct.job import Job
-from lekin.lekin_struct.machine import Machine
-from lekin.lekin_struct.operation import Operation
+from lekin.lekin_struct.machine import Machine  # 机器
+from lekin.lekin_struct.operation import Operation  # 工序
+from lekin.lekin_struct.order import Order
 from lekin.lekin_struct.route import Route

lekin/lekin_struct/material.py

@@ -0,0 +1,4 @@
+"""
+物料检查
+- 物料可以锁定，可以占用，可以空闲
+"""

lekin/solver/dispatching_rules/__init__.py → lekin/solver/construction_heuristics/__init__.py

@@ -2,5 +2,5 @@
 
 
 class Rules(object):
-    def __init__(self, name):
+    def __init__(self):
         pass

lekin/solver/construction_heuristics/core.py

@@ -0,0 +1,12 @@
+class Rules(object):
+    def __init__(self, name):
+        pass
+
+    def run(self):
+        pass
+
+    def _init(self):
+        return
+
+    def _update(self):
+        return

lekin/solver/construction_heuristics/epst.py

@@ -0,0 +1,11 @@
+"""Earliest Possible Start Time
+Forward scheduler
+正排"""
+
+
+class EPST(object):
+    def __init__(self):
+        pass
+
+    def solve(self):
+        pass

lekin/solver/construction_heuristics/lpst.py

@@ -0,0 +1,11 @@
+"""Latest Possible STart Time
+Backward scheduler
+倒排"""
+
+
+class LPST(object):
+    def __init__(self):
+        pass
+
+    def solve(self):
+        pass

lekin/solver/dispatching_rules/spt.py → lekin/solver/construction_heuristics/spt.py

@@ -1,11 +1,12 @@
 """Shortest Processing Time"""
 
+from collections import OrderedDict
 from typing import Any, Callable, Dict, List, Optional, Tuple, Union
 
-# from collections import sorteddict
-
 
 class SPT(object):
+    """ """
+
     def __init__(self):
         self.time = {}
         self.waiting_operations = {}
@@ -64,7 +65,7 @@ def solve(self, job_list: List, machine_list: List):
 
                 del self.time[t]
                 break
-            self.time = SortedDict(self.time)
+            self.time = OrderedDict(self.time)
         return self.jobs_list_to_export
 
     def get_waiting_operations(self, job_list, time, machine_list, current_time_on_machines):

lekin/solver/meta_heuristics/critical_path.py

@@ -0,0 +1,6 @@
+"""Critical path"""
+
+
+class CriticalPath(object):
+    def __init__(self):
+        pass

lekin/solver/meta_heuristics/hill_climbing.py

@@ -0,0 +1,6 @@
+"""Hill climbing"""
+
+
+class HillClimbing(object):
+    def __init__(self):
+        pass

lekin/solver/meta_heuristics/tabu_search.py

@@ -0,0 +1,9 @@
+"""
+Local search
+https://towardsdatascience.com/optimization-techniques-tabu-search-36f197ef8e25
+"""
+
+
+class TabuSearch(object):
+    def __init__(self):
+        pass

lekin/solver/meta_heuristics/variable_neighborhood_search.py

@@ -1 +1 @@
-"""variable neighborhood search"""
+"""Variable neighborhood search"""

lekin/solver/reinforcement_learning/models/graphcnn.py

@@ -0,0 +1,19 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class MLP(nn.Module):
+    def __init__(self, num_layers, input_dim, hidden_dim, output_dim):
+        super().__init__()
+
+    def forward(self, x):
+        return
+
+
+class GraphCNN(nn.Module):
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, x, graph_pool, adj):
+        return