Python3实现单级库存仿真,single echelon inventory assessment

本文主要是介绍Python3实现单级库存仿真,single echelon inventory assessment,希望对大家解决编程问题提供一定的参考价值,需要的开发者们随着小编来一起学习吧!

参考代码的来源:
https://github.com/anshul-musing/single-echelon-inventory-assessment/blob/master/src/simpy_3.0/simLostSales.py

src/simpy_3.0/simBackorder.py

这段代码主要模拟单级供应链,所考虑的库存参数为在途库存、库存水平、服务水平。
假设这个系统采用的是“一旦库存水平低于再订货点(固定),管理者立即下订单(固定)”的订货策略。
假设当前未被满足的订单允许被后期的补货满足,
基于订单有多晚被满足 ,计算服务水平。
假设需求服从正态分布、提前期服从均匀分布。

"""This module simulates a single-echelon supply chain
and calculates inventory profile (along with associated inventory
parameters such as on-hand, inventory position, service level, etc.)
across timeThe system follows a reorder point-reorder quantity policy
If inventory position <= ROP, an order of a fixed reorder
quantity (ROQ) is placed by the facilityIt is assumed that any unfulfilled order is backordered
and is fulfilled whenever the material is available in the
inventory.  The service level is estimated based on how
late the order was fulfilledDemand is assumed to be Normally distributed
Lead time is assumed to follow a uniform distribution
"""__author__ = 'Anshul Agarwal'import simpy
import numpy as np# Stocking facility class
class stockingFacility(object): ## ?? why we need to in herit 'object'?# initialize the new facility objectdef __init__(self, env, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime):self.env = envself.on_hand_inventory = initialInvself.inventory_position = initialInvself.ROP = ROP # inventory positionself.ROQ = ROQ # fixed order quantityself.meanDemand = meanDemandself.demandStdDev = demandStdDevself.minLeadTime = minLeadTimeself.maxLeadTime = maxLeadTimeself.totalDemand = 0.0self.totalBackOrder = 0.0self.totalLateSales = 0.0self.serviceLevel = 0.0env.process(self.runOperation())# main subroutine for facility operation# it records all stocking metrics for the facilitydef runOperation(self):while True:yield self.env.timeout(1.0)# demand newly generateddemand = float(np.random.normal(self.meanDemand, self.demandStdDev, 1))self.totalDemand += demand# shipment 是该仓库送出的量,而self.ROQ是该仓库的补货量shipment = min(demand + self.totalBackOrder, self.on_hand_inventory) # the amount of goods available to sendself.on_hand_inventory -= shipment # send the shipment to some retailerself.inventory_position -= shipmentbackorder = demand - shipment # the amount of demand unmet temporarilyself.totalBackOrder += backorderself.totalLateSales += max(0.0, backorder)# if the current inventory position is less than ROP, then place an orderif self.inventory_position <= 1.01 * self.ROP:  # multiply by 1.01 to avoid rounding issuesself.env.process(self.ship(self.ROQ))# why we revise 'self.on_hand_inv' in the method 'ship', and revise 'self.inv_position' outside 'ship'self.inventory_position += self.ROQ# subroutine for a new order placed by the facilitydef ship(self, orderQty):# recall that we assume the lead time follows an uniform distributionleadTime = int(np.random.uniform(self.minLeadTime, self.maxLeadTime, 1))yield self.env.timeout(leadTime)  # wait for the lead time before delivering# now 'orderQty' goods is receivedself.on_hand_inventory += orderQty# Simulation module
def simulateNetwork(seedinit, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime):env = simpy.Environment()  # initialize SimPy simulation instancenp.random.seed(seedinit)s = stockingFacility(env, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime)env.run(until=365)  # simulate for 1 years.serviceLevel = 1 - s.totalLateSales / s.totalDemand # 服务水平的定义:那些被及时满足的需求的占比return s######## Main statements to call simulation ########
meanDemand = 500.0
demandStdDev = 100.0
minLeadTime = 7
maxLeadTime = 13
CS = 5000.0
ROQ = 6000.0
ROP = max(CS,ROQ)
initialInv = ROP + ROQ# Simulate
replications = 100
sL = []
for i in range(replications):nodes = simulateNetwork(i,initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime)sL.append(nodes.serviceLevel)sLevel = np.array(sL)
print("Avg. service level: " + str(np.mean(sLevel)))
print("Service level standard deviation: " + str(np.std(sLevel)))

src/simpy_3.0/simLostSales.py

不同于上一小节的地方在于,这里不允许回购,而是允许发生销售损失(lost sales)。
因此,在代码实现方面也会有微妙的差别,具体如下,

  1. 在类stockingFacility中数据self.totalShipped用于记录从这个仓库发出了多少货;
  2. 在类stockingFacility的方法runOperation中,当前从该仓库的送出量shipment的计算方式不再考虑backorder;
  3. 在函数simulateNetwork中,计算服务水平(从该仓库的送出量占总需求量的比例)。
"""This module simulates a single-echelon supply chain
and calculates inventory profile (along with associated inventory
parameters such as on-hand, inventory position, service level, etc.)
across timeThe system follows a reorder point-reorder quantity policy
If inventory position <= ROP, an order of a fixed reorder
quantity (ROQ) is placed by the facilityIt is assumed that any unfulfilled order is lost
The service level is estimated based on how much
of the demand was fulfilledDemand is assumed to be Normally distributed
Lead time is assumed to follow a uniform distribution
"""__author__ = 'Anshul Agarwal'import simpy
import numpy as np# Stocking facility class
class stockingFacility(object):# initialize the new facility objectdef __init__(self, env, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime):self.env = envself.on_hand_inventory = initialInvself.inventory_position = initialInvself.ROP = ROPself.ROQ = ROQself.meanDemand = meanDemandself.demandStdDev = demandStdDevself.minLeadTime = minLeadTimeself.maxLeadTime = maxLeadTimeself.totalDemand = 0.0self.totalShipped = 0.0 # !!self.serviceLevel = 0.0env.process(self.runOperation())# main subroutine for facility operation# it records all stocking metrics for the facilitydef runOperation(self):while True:yield self.env.timeout(1.0)demand = float(np.random.normal(self.meanDemand, self.demandStdDev, 1))self.totalDemand += demandshipment = min(demand, self.on_hand_inventory) # !!self.totalShipped += shipmentself.on_hand_inventory -= shipmentself.inventory_position -= shipmentif self.inventory_position <= 1.01 * self.ROP:  # multiply by 1.01 to avoid rounding issuesself.env.process(self.ship(self.ROQ))self.inventory_position += self.ROQ# subroutine for a new order placed by the facilitydef ship(self, orderQty):leadTime = int(np.random.uniform(self.minLeadTime, self.maxLeadTime, 1))yield self.env.timeout(leadTime)  # wait for the lead time before deliveringself.on_hand_inventory += orderQty# Simulation module
def simulateNetwork(seedinit, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime):env = simpy.Environment()  # initialize SimPy simulation instancenp.random.seed(seedinit)s = stockingFacility(env, initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime)env.run(until=365)  # simulate for 1 years.serviceLevel = s.totalShipped / s.totalDemand # !!return s######## Main statements to call simulation ########
meanDemand = 500.0
demandStdDev = 100.0
minLeadTime = 7
maxLeadTime = 13
CS = 5000.0
ROQ = 6000.0
ROP = max(CS,ROQ)
initialInv = ROP + ROQ# Simulate
replications = 100
sL = []
for i in range(replications):nodes = simulateNetwork(i,initialInv, ROP, ROQ, meanDemand, demandStdDev, minLeadTime, maxLeadTime)sL.append(nodes.serviceLevel)sLevel = np.array(sL)
print("Avg. service level: " + str(np.mean(sLevel)))
print("Service level standard deviation: " + str(np.std(sLevel)))

这篇关于Python3实现单级库存仿真,single echelon inventory assessment的文章就介绍到这儿,希望我们推荐的文章对编程师们有所帮助!



http://www.chinasem.cn/article/906506

相关文章

pandas中位数填充空值的实现示例

《pandas中位数填充空值的实现示例》中位数填充是一种简单而有效的方法,用于填充数据集中缺失的值,本文就来介绍一下pandas中位数填充空值的实现,具有一定的参考价值,感兴趣的可以了解一下... 目录什么是中位数填充?为什么选择中位数填充?示例数据结果分析完整代码总结在数据分析和机器学习过程中,处理缺失数

Golang HashMap实现原理解析

《GolangHashMap实现原理解析》HashMap是一种基于哈希表实现的键值对存储结构,它通过哈希函数将键映射到数组的索引位置,支持高效的插入、查找和删除操作,:本文主要介绍GolangH... 目录HashMap是一种基于哈希表实现的键值对存储结构,它通过哈希函数将键映射到数组的索引位置,支持

Pandas使用AdaBoost进行分类的实现

《Pandas使用AdaBoost进行分类的实现》Pandas和AdaBoost分类算法,可以高效地进行数据预处理和分类任务,本文主要介绍了Pandas使用AdaBoost进行分类的实现,具有一定的参... 目录什么是 AdaBoost?使用 AdaBoost 的步骤安装必要的库步骤一:数据准备步骤二:模型

使用Pandas进行均值填充的实现

《使用Pandas进行均值填充的实现》缺失数据(NaN值)是一个常见的问题,我们可以通过多种方法来处理缺失数据,其中一种常用的方法是均值填充,本文主要介绍了使用Pandas进行均值填充的实现,感兴趣的... 目录什么是均值填充?为什么选择均值填充?均值填充的步骤实际代码示例总结在数据分析和处理过程中,缺失数

Java对象转换的实现方式汇总

《Java对象转换的实现方式汇总》:本文主要介绍Java对象转换的多种实现方式,本文通过实例代码给大家介绍的非常详细,对大家的学习或工作具有一定的参考借鉴价值,需要的朋友参考下吧... 目录Java对象转换的多种实现方式1. 手动映射(Manual Mapping)2. Builder模式3. 工具类辅助映

Go语言开发实现查询IP信息的MCP服务器

《Go语言开发实现查询IP信息的MCP服务器》随着MCP的快速普及和广泛应用,MCP服务器也层出不穷,本文将详细介绍如何在Go语言中使用go-mcp库来开发一个查询IP信息的MCP... 目录前言mcp-ip-geo 服务器目录结构说明查询 IP 信息功能实现工具实现工具管理查询单个 IP 信息工具的实现服

SpringBoot基于配置实现短信服务策略的动态切换

《SpringBoot基于配置实现短信服务策略的动态切换》这篇文章主要为大家详细介绍了SpringBoot在接入多个短信服务商(如阿里云、腾讯云、华为云)后,如何根据配置或环境切换使用不同的服务商,需... 目录目标功能示例配置(application.yml)配置类绑定短信发送策略接口示例:阿里云 & 腾

python实现svg图片转换为png和gif

《python实现svg图片转换为png和gif》这篇文章主要为大家详细介绍了python如何实现将svg图片格式转换为png和gif,文中的示例代码讲解详细,感兴趣的小伙伴可以跟随小编一起学习一下... 目录python实现svg图片转换为png和gifpython实现图片格式之间的相互转换延展:基于Py

Python利用ElementTree实现快速解析XML文件

《Python利用ElementTree实现快速解析XML文件》ElementTree是Python标准库的一部分,而且是Python标准库中用于解析和操作XML数据的模块,下面小编就来和大家详细讲讲... 目录一、XML文件解析到底有多重要二、ElementTree快速入门1. 加载XML的两种方式2.

Java的栈与队列实现代码解析

《Java的栈与队列实现代码解析》栈是常见的线性数据结构,栈的特点是以先进后出的形式,后进先出,先进后出,分为栈底和栈顶,栈应用于内存的分配,表达式求值,存储临时的数据和方法的调用等,本文给大家介绍J... 目录栈的概念(Stack)栈的实现代码队列(Queue)模拟实现队列(双链表实现)循环队列(循环数组