本文为APS高级计划与排程专题系列文章一：什么是APS高级计划与排程？
一、APS高级计划与排程
APS高级计划与排程是什么？什么是APS高级计划与排程？APS高级计划与排程(Advanced Planning and Scheduling)，是解决生产排程和生产调度问题，常被称为排序问题或资源分配问题。
在离散行业，APS是为解决多工序、多资源的优化调度问题。数字化转型网szhzxw.cn
在流程行业，APS则是为解决顺序优化问题。
APS通过为流程和离散的混合模型同时解决顺序和调度的优化问题，从而对项目管理与项目制造解决关键链和成本时间最小化，具有重要意义。数字化转型网szhzxw.cn
二、APS的起源与发展
从20世纪40年代以来，用数学方法进行精确计算来安排生产计划,就一直是一个传统的研究课题。高级计划排程的一些主要思想早在计算机的存在前就已经出现。对 APS 贡献最大的有两个方面：一是早在 20 世纪初出现的甘特图( Gantt Chart) ；二是运用数学规划模型解决计划问题。美国和前苏联都应用新的最优化线性规划技术解决与战争相关的物流管理问题。这些思想和方法对于 APS的萌芽起到了奠基性的作用。
随着跨国公司在世界各地的发展 ，制造业问题变得越来越复杂 ,变量规模达到数以万计。虽然线性规划等技术也扩展成可以处理更加复杂的问题 ，但仍然不能满足企业的需要。因此 ，许多公司在内部开发自己的APS,另一些公司则在购买解决线性规划问题的程序基础上进行开发。数字化转型网szhzxw.cn
60 年代中期 , IBM 开发了基于产品结构分解的MRP系统，并在 70 年代发展为闭环 MRP系统 ,除了物料需求计划外 ，还将生产能力需求计划、车间作业计划和采购作业计划也全部纳入 MRP ，形成一个封闭的系统。这为 80 年代 MRPII的出现奠定了基础，但实际上MRPII的这种闭环因是预设提前期、无限制的产能计划排产与无约束的物料计划，故只能是手工闭环，难以匹配实际复杂动态的制造环境。
这段时间 ,模拟技术开始进入计划领域，基于模拟的计划工具开始出现；而到了80年代初 , 轮胎制造商Kelly Springfield 和烟草公司 Philip Morris 开始应用计划和排程系统。随后快速MRP的模拟技术，能将复杂的生产作业模拟在独立计算机上,部分采用以常驻内存方式进行批处理运算，脱离了当时占业务计算支配地位的主机，使制造企业完成生产计划排程只用几小时而不是当时所公认的20多个小时，大大缩短了计划运行时间。 数字化转型网szhzxw.cn
1984年AT&T推出的 Karmarkar’s 算法，成为线性规划突破性进展，这个新技术解决了线性规划的问题，是AT&T作为“真正的突破”和“设计解决了以前未解决的问题”。AT&T把这个算法绑定他们的计算机，价格达到了天价般的九百万美元！APS成为新兴企业管理者的宠儿。许多大型化工公司巴斯夫等都开始积极使用计划和排程的工具， 甚至连许多大的航空公司如美航也实施了复杂的计划和排程系统。
而数据库技术SQL的引进，允许APS工具和关系型数据库更动态的互动。APS开始攻城略地了。1990年初,消费品公司(CPG)开始引入APS系统，他们需要更复杂的系统；电子装配、金属品制造等离散制造领域被i2、Fastman打开洞门，蜂拥而进入。半导体领域如IBM、Intel、TI、Harris公司等成为APS发展的重要推手。其中i2公司以彪悍的市场导向和销售战略，戏剧性地提高了APS的发展空间。数字化转型网szhzxw.cn
随着APS引擎的成熟，使理论化的数学解析计划方法达到了实用程度，生产计划方法交替，ERP也出现了继续完善和功能扩充以及改变ERP的功能和性质这两种发展趋势。随着APS市场的快速成长，产生了ERP供应商的新一轮的收购APS公司或自己内部开发APS。
许多专家认为APS必须嵌入ERP系统，一些厂商也开发了APS模块。近些年来，几乎每个外国的ERP软件，都有了与之集成的APS引擎。如SAP 有了 SAP APO (Advanced Planning and Optimization), Baan 有Baan SCS (Supply Chain Solutions), Oracle 买了ILOG的产品, PeopleSoft 购买了Red Pepper，JD Edwards 买了Numetrix，ERP和APS正在融合在一起。同样，MES也在把APS纳入自己的制造板块，如GE 收购了合作伙伴Novotech的Scheduler, 西门子收购了Preactor，钢铁行业MES 巨头PSI兼并了钢铁行业的APS Broner。
APS 的主要目标是某一指标(例如 设备资产回报率ROA) 的总优化。这需要预先假定供应链的结构和其各种不同的资源和边界情况(例如能力限制)必须是可见的。实际上，APS 工具是以供应链的跨组织模型为基础的。这也是在数字化制造下，APS越加显得耀眼的原因。数字化转型网szhzxw.cn
三、市场上的APS软件数字化转型网szhzxw.cn
从上个世纪初开始，是APS在中国的概念推广阶段，一些高校研究者将APS的理论从高校实验室里搬到企业应用现场，也有海外的APS实践者们纷纷回到国内进行APS应用推广，同时值得关注的还有一些实践APS的国内企业先行者。这三类群体构成了当前国内APS热潮中主流力量。说到优化算法和排产软件，就不得不提大名鼎鼎的ILOG（现隶属于IBM旗下）和大名鼎鼎的Cplex。Cplex是IBM公司一款高性能的数学规划问题求解器。Cplex被许多公司做为APS的核心引擎，如ILOG的 PPO，SAP的APO，i2的优化软件。此外国外APS品牌有Asprova、FlexSche 、Preactor（西门子）、AspenTeech、Quintiq等。而国内有施达优、安达发、兰光创新、永凯、元工国际等产品。数字化转型网szhzxw.cn
但也不得不提的尴尬的是，APS也没有太大的公司，或者在它刚要长大的时候，就被ERP厂商哐当一口吃掉。想来要独立发展的专业软件，都会面临着这般的成长烦恼。
四、APS的趋势与展望
APS的核心是优化算法，已经走过四代。最新第四代是智能算法融合人工智能动态调整算法，以智能算法进行静态排程，以多Agent代理协商进行分布计算动态调整。迄今为止，以MPS和MRP运算为核心思想的计划管理已经成为现代企业ERP的标准和核心功能，但是从几十年的应用效果看，依然难以满足企业的计划管理需求。其实MPS主生产计划和MRP物料需求计划的体系与方式，已很难适应按需生产环境,尤其无法适应大规模个性化定制的工业4.0时代。而APS可以综合考虑产能、工装、加工批次等约束，并能结合MES实现滚动排产。数字化转型网szhzxw.cn
这也说明了，为什么在工业4.0时代，MES和APS系统，成为比ERP要耀眼得多的明星。APS分为供应链级的APS和工厂级的APS两类。而制造执行系统MES，最早期用于车间控制、管理工作中心,并下达派工单，主要靠手工汇报进度和工时；随着制造精益化，制造数字化，形成了MES制造执行系统。因此二者在智能排产功能是有重叠的。当下作为代表当今先进管理思想的高级计划排程系统，APS的核心是久经磨练的数学算法或解决方案。在数字化经济，APS的发展呈现多元化的趋势一个是与ERP、MES更加紧密结合。
值得注意的是，APS的数据来源正在发生变革。APS数据如i2，Oracle的APS，都源自于ERP；而随着MES的普及，生产系统同样成为一个巨大的数据池源发地。这对于APS是一个全新的发展机遇。一个是与多品种小批量订单制造和项目制造结合：实际上，只有APS才能实现多品种小批量的个性化定制计划模式。同样值得注意的是云平台化，由于APS的服务的间歇性，造成购买的高昂的核心算法和服务器运行时间的闲置，为此，将APS云平台化可以大大降低企业的计划排产的投资，国内外已经有企业在部署SaaS的APS。
尽管APS具有强大的功能，非常适合于供应链整体计划问题的解决，但在我国企业管理实际中，APS应用仍然存在很多需要解决的问题。例如APS功能的发挥主要在供应链管理上，而我国企业的供应链管理还停留在非常初级的阶段，企业对于供应链竞争的意识还很模糊。数字化转型网szhzxw.cn
因此，APS难以找到真正的用武之地。而企业内部信息化基础数据、流程、计划体系薄弱。APS的运行需要有ERP、MES、PLM等系统提供的数据支持。此前，很多APS项目实施效果不好的原因大多是因为没有MES帮助APS实现闭环和滚动排产，导致计划兑现率较差。尽管如此，随着市场针对大规模个性化定制的需求不断的增加，APS已经逐渐成为离散制造行业智能工厂的重要中枢。
在一些行业，如按照分组加工的烟草行业、大规模个性化生产的汽车和家电行业，如果没有APS，人工已经很难基于经验进行计划的编制工作。随着，个性化需求逐渐成为市场主流需求，预计未来三到五年内，APS 在我国的应用将开始进入大量实施阶段，成为智能工厂的指挥中心。数字化转型网szhzxw.cn
更多APS高级计划与排程相关文章、APS高级计划与排程相关案例、APS高级计划与排程相关方案，可在数字化转型网szhzxw.cnAPS高级计划与排程栏目进行查看；APS高级计划与排程相关数字化服务商可在数字化转型网szhzxw.cn找方案-APS高级计划与排程栏目进行查看；如需获取更多APS高级计划与排程相关资料包，可点击蓝字联系我们获取；也可扫码加入读者社群与我们一同探讨~
APS高级计划与排程专题系列文章（一）：什么是APS高级计划与排程？翻译：数字化转型网szhzxw.cn
This article is part 1 of the APS Advanced Planning and Scheduling Series: What is APS Advanced Planning and Scheduling?
I. Advanced APS planning and scheduling数字化转型网szhzxw.cn
What is APS Advanced Planning and Scheduling? What is APS Advanced Planning and Scheduling? APS Advanced Planning and Scheduling solves production scheduling and scheduling problems, often referred to as scheduling problems or resource allocation problems.
In discrete industries, APS is designed to solve the optimization scheduling problem of multi-processes and multi-resources.数字化转型网szhzxw.cn
In the process industry, APS is designed to solve sequential optimization problems.
APS has important implications for project management and project manufacturing by addressing both sequential and scheduling optimization issues for process and discrete hybrid models, thereby minimizing critical chains and cost time.
Ii. Origin and development of APS
The use of mathematical methods to make precise calculations to schedule production has been a traditional research topic since the 1940s. Some of the main ideas for advanced scheduling predate computers. The biggest contribution to APS comes from two aspects: the Gantt Chart, which appeared in the early 20th century; The second is to use mathematical programming model to solve the planning problem. Both the United States and the former Soviet Union applied new optimized linear programming techniques to solve logistics management problems related to war. These ideas and methods have played a fundamental role in the germination of APS.数字化转型网szhzxw.cn
With the development of multinational corporations around the world, manufacturing issues become more and more complex, with the scale of variables running into tens of thousands. Although techniques such as linear programming have been expanded to deal with more complex problems, they still do not meet the needs of enterprises. As a result, many companies develop their APS in-house, while others develop them on the basis of purchasing programs that solve linear programming problems.
数字化转型网szhzxw.cn
In the mid-1960s, IBM developed the MRP system based on product structure decomposition, and in the 1970s, it developed the closed-loop MRP system. In addition to material requirement planning, it also incorporated the production capacity requirement planning, shop operation planning and procurement operation planning into MRP, forming a closed system. This laid the foundation for the emergence of MRPII in the 1980s. But in fact, the closed-loop of MRPII can only be a manual closed-loop, which is difficult to match the actual complex and dynamic manufacturing environment, due to the preset lead time, unlimited capacity planning and unconstrained material planning.
During this period, simulation techniques began to enter the planning field and planning tools based on simulation began to appear. By the early 1980s, Kelly Springfield, a tyremaker, and Philip Morris, a tobacco firm, were using planning and scheduling systems. Subsequently, the simulation technology of fast MRP can simulate complex production operations on an independent computer, and part of it adopts the resident memory mode for batch processing, which is separated from the dominant business computing host at that time, so that the manufacturing enterprises can complete the production schedule only in a few hours instead of more than 20 hours which was recognized at that time, greatly shortening the planned running time.
数字化转型网szhzxw.cn
In 1984, AT&T launched Karmarkar’s algorithm, which became a breakthrough in linear programming. This new technology solved the problem of linear programming, and it was AT&T as a “real breakthrough” and “design to solve previously unsolved problems”. AT&T bundled the algorithm into their computers for a whopping $9 million! APS became the darling of start-up managers. Many large chemical companies, such as BASF, have begun to actively use planning and scheduling tools, and even large airlines, such as American Airlines, have implemented sophisticated planning and scheduling systems.
The introduction of database technology SQL allows APS tools to interact more dynamically with relational databases. The APS is starting to gain ground. In early 1990, consumer products companies (CPGS) began to introduce APS systems. They needed more complex systems; i2 and Fastman rushed into discrete manufacturing sectors such as electronic assembly and metal manufacturing. Semiconductor companies such as IBM, Intel, TI and Harris have become important drivers of APS development. Among them, i2 company dramatically improved the development space of APS with its fierce market orientation and sales strategy.
数字化转型网szhzxw.cn
With the maturation of APS engine, the theoretical mathematical analytic planning method has reached the practical level, the production planning method is alternating, ERP also appears to continue to improve and function expansion, and change the function and nature of ERP. With the rapid growth of APS market, ERP suppliers are acquiring APS companies or developing APS in-house.
Many experts believe that APS must be embedded in ERP system, and some vendors have developed APS modules. In recent years, almost every foreign ERP software has an APS engine integrated with it. For example, SAP has SAP APO (Advanced Planning and Optimization). Baan has Baan SCS (Supply Chain Solutions), Oracle has bought ILOG products, PeopleSoft bought Red Pepper, JD Edwards bought Numetrix, and ERP and APS are merging. Similarly, MES is also incorporating APS into its manufacturing segment. For example, GE acquired Scheduler of its partner Novotech, Siemens acquired Preactor, and PSI, MES giant of steel industry, acquired APS Broner of steel industry.数字化转型网szhzxw.cn
The primary goal of APS is the total optimization of a metric such as ROA. This presuppositions that the structure of the supply chain and its various resource and boundary conditions (such as capacity constraints) must be visible. In fact, APS tools are based on a cross-organizational model of the supply chain. This is also the reason why APS is increasingly dazzling in digital manufacturing.
Third, APS software in the market数字化转型网szhzxw.cn
At the beginning of the last century, it was the concept promotion stage of APS in China. Some university researchers moved the theory of APS from the university laboratory to the enterprise application site, and overseas APS practitioners returned to China to promote the application of APS. Meanwhile, it is worth paying attention to some domestic enterprise pioneers who practiced APS.
These three groups constitute the mainstream force in the current APS boom in China. When it comes to optimization algorithms and scheduling software, there’s ILOG (now owned by IBM) and Cplex. Cplex is a high-performance mathematical programming problem solver from IBM. Cplex is used as the core engine of APS by many companies, such as ILOG’s PPO, SAP’s APO, and i2 optimization software. In addition, foreign APS brands include Asprova, FlexSche, Preactor (Siemens), AspenTeech, Quintiq, etc. In China, there are Shi Da You, Anda Fa, LAN Guang Innovation, Yongkai, Yuangong International and other products.数字化转型网szhzxw.cn
However, it is also embarrassing to mention that APS is not a big company, or when it was just about to grow up, it was eaten by ERP manufacturers. Professional software that wants to stand on its own faces such growing pains.
Trend and prospect of APS数字化转型网szhzxw.cn
At its core, APS is an optimization algorithm that has been around for four generations. The latest fourth generation is the integration of intelligent algorithm and artificial intelligence dynamic adjustment algorithm. Using intelligent algorithm for static scheduling, using multi-agent Agent negotiation for distributed computing dynamic adjustment. So far, the plan management with MPS and MRP operation as the core idea has become the standard and core function of modern enterprise ERP. But from the decades of application effect, it is still difficult to meet the needs of enterprise plan management. In fact, the system and mode of MPS master production plan and MRP material requirement plan have been difficult to adapt to the on-demand production environment. Especially the industry 4.0 era of mass customization. APS can comprehensively consider constraints such as capacity, tooling and processing batches, and realize rolling production scheduling in combination with MES.数字化转型网szhzxw.cn
数字化转型网szhzxw.cn
This also explains why in the era of industry 4.0, MES and APS systems have become much more dazzling stars than ERP. APS is classified into supply chain APS and factory APS. Manufacturing execution system MES, the earliest used for workshop control, management of the work center, and dispatch orders, mainly rely on manual reporting progress and working hours. With lean manufacturing and digital manufacturing, MES manufacturing execution system has been formed. Therefore, there is overlap between the two functions of intelligent scheduling. As an advanced scheduling system representing today’s advanced management ideas, the core of APS is a well-honed mathematical algorithm or solution. In the digital economy, the development of APS presents the trend of diversification. One is to combine more closely with ERP and MES.数字化转型网szhzxw.cn
数字化转型网szhzxw.cn
It is worth noting that the source of APS data is changing. APS data such as i2 and Oracle APS are derived from ERP; With the popularization of MES, the production system also becomes a huge source of data pool. This is a new development opportunity for APS. One is to combine with multi-variety small-batch order manufacturing and project manufacturing: in fact. Only APS can realize the multi-variety small-batch personalized customized planning mode. It is also worth noting that the cloud platform, due to the intermittently APS service, resulting in the purchase of expensive core algorithms and server running time idle, therefore. APS cloud platform can greatly reduce the enterprise’s investment in production planning, domestic and foreign enterprises have deployed SaaS APS.数字化转型网szhzxw.cn
Although APS has powerful functions and is very suitable for the solution of the overall supply chain planning problem, there are still many problems to be solved in the application of APS in Chinese enterprise management practice. For example, the application of APS function is mainly in the supply chain management. However, the supply chain management of Chinese enterprises remains at a very primary stage.A and the enterprise’s consciousness about the supply chain competition is vague.
数字化转型网szhzxw.cn
As a result, APS has struggled to find a real use. And enterprise internal information basic data, process, planning system is weak. APS operation requires data support from ERP, MES, PLM and other systems. Previously, many APS projects had poor implementation results mostly because there was no MES to help APS realize closed loop and rolling production scheduling, leading to poor planned cash rate. However, with the increasing demand for mass customization, APS has become an important hub for smart factories in discrete manufacturing industries.
数字化转型网szhzxw.cn
In some industries, such as tobacco. Which is processed in groupings, and automobiles and home appliances, which are manufactured in mass personalization, it is already difficult for people to make plans based on experience without APS. With personalized demand gradually becoming the mainstream demand of the market. It is expected that the application of APS in our country will begin to enter a large number of implementation stages in the next three to five years and become command center of intelligent factories.数字化转型网szhzxw.cn
数字化转型网szhzxw.cn
More APS advanced planning and scheduling related articles. APS advanced planning and scheduling related cases, APS advanced planning and scheduling related solutions. Can be viewed in the column of Digital transformation website szhzxw.cnAPS Advanced Planning and Scheduling. APS Advanced Planning and scheduling related digital service providers can find solutions on the digital transformation network szhzxw.cn -APS Advanced Planning and Scheduling column for viewing; For more APS Advanced Planning and Scheduling related information packs, please contact us by clicking on the blue letters
本文由数字化转型网（www.szhzxw.cn）转载而成，来源：百度百科；编辑/翻译：数字化转型网默然。
免责声明: 本网站(http://www.szhzxw.cn/)内容主要来自原创、合作媒体供稿和第三方投稿，凡在本网站出现的信息，均仅供参考。本网站将尽力确保所提供信息的准确性及可靠性，但不保证有关资料的准确性及可靠性，读者在使用前请进一步核实，并对任何自主决定的行为负责。本网站对有关资料所引致的错误、不确或遗漏，概不负任何法律责任。
本网站刊载的所有内容(包括但不仅限文字、图片、LOGO、音频、视频、软件、程序等) 版权归原作者所有。任何单位或个人认为本网站中的内容可能涉嫌侵犯其知识产权或存在不实内容时，请及时通知本站，予以删除。
