In this chapter of the Simple Supply Chain series, I look at the implications and impact of computer systems in the supply chain. There are many different options to choose from, but I decided to concentrate on ERP and MRP systems since they essentially present a comprehensive approach rather than a singular focused approach. Whatever system you face, just know that at the core it is extremely simple.
Computers – The Stupid Machine
A computer is used to help process information quickly. It is a stupid machine in that it only does what you tell it to do, as such it is no more intelligent or capable than a complex plumbing system full of faucets, gauges, filters and what have you. A computer process data, while plumbing systems process water. So, whenever anyone blames the computer, what they mean to say is that they are too stupid to input the data properly. Similar to someone complaining that there is only cold water when they don’t turn open the hot water tap, or forget to turn the heating system on.
Supply Chain IT
What is Supply Chain IT? Where does it begin and where does it end?
Well, Supply Chain IT is no different from any other IT system, the software used to supplement supply chains come in four forms:
- Storing Data
- Processing Data
- Analyzing Data
- Presenting Data
These are the four core processes that any IT system will perform. Let take a look at each one with respect to Supply Chain Systems.
This is all about creating databases of information. A database is as the name supposes, a base of data, like a library, but in its most basic form. A database is a field that has a title, a characteristic and a content. The title or name is the identifier that the software will use to identify a specific set of data or individual data within the set. A characteristic is the type of data being stored, this can be a number, a text, a combination of all, it can be bit or a number of characters dependant on the size of the field, it can be a single logical expression like yes/no, or it can be a string of words with unlimited size, it can even be a link, a picture, a video or whatever digital data format you want it to be. The content is what you place in the field.
Databases are made of rows and columns, all databases (except for (SAP Hana) manage databases where the row is the number of items in a data set, and the columns are the fields that make up a data set.
A database at its core is basically a list, it can be in Excel, in a word, in the text, it can be in a complex ERP system like SAP Hana, or in a blockchain. Databases are the libraries of the software world, and they are how you store information for retrieval.
Some database have master data files; these are databases of static information that is not expected to change frequently, such as a products material, weight, dimensions, or suppliers, name and address and contact information. Master data files are databases of information that are considered to be the core of a database system, and everything else is fluid around it, such as generating a purchase order, which is reliant on master data information, but the order itself is unique and individual.
Processing data is how you take the information you placed in the database, and perform specific tasks. Processing is not just fulfilling equations or processing algorithms, processing includes transactions between data sets, for instance: You wish to make a purchase order, you have a database of products, you have the minimum order quantity set, and the stakeholder demands 50 items more than is in the system, even when the minimum order quantity is not yet reached. This request is generated by a demand that came in from sales for an increase in production. The inputted request generates a process where the system takes the information and creates a message. The message is made to look like a request (presenting data), it also takes all the information regarding the parts to be ordered, such as price, supplier, lead delivery times and whatever is relevant. This is then passed to an authorizing agent and once authorized the system processes the authorization, opening up a purchase order. The purchase order is now sent to the supplier.
So far, the system has not performed any deep analysis; it has only reacted to a physical request, not an automated one. Processing data is what happens all the time, and every time the system is accessed for information.
The core difference between analyzing data and processing it is that analysis is about passing data through an algorithm to produce a result. For instance, a process of data is taking data and looking up the relevant components in one or more datasets, and creating a new dataset. The analysis will take information and pass it through algorithms and functions that are designed to find patterns, trends, statistics, mathematical and engineering formulations, and provide results and present information that went through complex calculations and not just dataset connections and other database related processing.
For instance, the minimum order quantity is a process, it is setting a limit to a data set, and when the system reaches the preset count, it will trigger a process.
The cumulative amount of orders that were processed in a year based on reaching minimum quantity levels under 30 days is an analysis that will give a result based on calculating a specific operation. The same is kind of analysis, but in a more complex set of calculations is calculating the trajectory and flight path required to fly a Tesla to Mars.
With all this processing and calculation, perhaps the most important of all the functions is presenting the data. This can be as simple as a screen view of fields, to a print out of a three-dimensional part in a 3D printer, to the plotting of the said flight path and trajectory.
Taking the data, and analyzing it will only be effective if you can present it to the human for decision-making processes.
The Complex is Simple
You might think that all of this complexity is mind-boggling and complex, but it isn’t, in fact, its one of the simplest parts of the industry. IT is not new, and it’s not mind-boggling, and in most cases, it is not about managing big data. 99% of the world is not Amazon or Google; it is companies that employ 10 people and struggle with the IRS.
In most cases, you don’t even need a server room, although having a cooled server room always helps for performance. It also separates the IT core from the mainstream.
To be simple, an IT core is a computer (can be a server, a server is just a dedicated computer to a specific client) that is connected using either LAN, WAN or internet and enables users to log in and interface with the system.
Think of it as a hotel with lots of rooms, you have your room key, and from your room, you can dial up room service, housekeeping, the front desk and more. The same goes with an ERP system, you get access to a specific module or modules, there are your hotel room keys, and you can perform specific tasks within the system as set down by the authorization that is allocated to your room key.
The ERP/MRP system has many users, and each one is in charge of a specific set of data. When everyone inputs their data and manages their processes, then the decision makers can analyze the data and reach informed decisions.
If you enter bad data, you will affect the entire system. As such, the importance of data maintenance is key to assure a steady flow of accurate data.
Supply Chain Hardware
One of the biggest changes in supply chain IT comes from the hardware. This means that a supply chain system not only has software managing materials, orders, and tracking materials, it also has a hand in actually handling materials. This is seen in automated warehouse systems, where automated conveyors are used, where automated robot storage units are integrated and where RFID and barcode scanners are used to collect data on the go.
Again, these might seem complex and to organize them in the correct configuration does require a degree in engineering, but essentially you are looking at it from an IT point of view, and this is where the software used to provide a base for inputting data is actually very simple.
It’s not all down to the data you input if it’s a conveyor speed or the way a robot arm moves in synchronicity with a conveyor. The actual figures are complex, but the software that manages it is basic. More so are the input (GUI) screens, where a good system will be simple to use, and a bad one will require the operator to have a degree in science. Graphic User Interface, GUI is the core of a successful IT (ERP/MRP) system.
One of the biggest arguments against SAP is that it’s not GUI friendly. A system that is too rigid, and where every screen change requires hours of calculations, describes a heavy system that is not designed to be adaptable to change. However, what makes SAP so great, is their marketing and PR. Oh, and the fact that their systems, while being very rigid, are actually quite comprehensive.
Open Source vs. Closed Systems
To be honest, an open source solution will always outperform a closed system. Open source software enables you to enjoy apps from many sources, with good results, providing you with a bespoke system designed for your specific needs. However, the only downside to OS systems is that the customer service is either nonexistent, or you need to remember that anyone and everyone can figure out how to access your data. A closed system will have customer support, will cost more money, but will be more secure.
Resource Allocation is the Key
An ERP/MRP system will provide you with excellent processing, analytics, and hands-on data when you need it. However, the data will only be as good as the allocation of resources is managed. In IT systems, resource allocation relates to three areas: the allocation of software to manage data, the allocation of the resources in the ERP system to specific budgets and the allocation of people to manage specific data sets.
Three completely different areas, all dealing in the same thing: resource allocation.
Simple but Effective
- Allocation of Software: if the right software is used, including accounting, sales, project management, materials management, warehouse management, production management, procurement and logistics (Supply Chain), with or without lean management and 6S and other specialist modules, any company with a supply chain will be able to provide a comprehensive easy to maintain dataset for processing, tracking, analytics and decision making.
- Allocation of Resources: This is all about the way you put in the data. You need to allocate the data correctly. A small mistake will ruin an entire system. Computers, even with AI will not realize you made a mistake and will process the mistake. For instance, you placed kilograms instead of pounds, or you deiced on a minimum order quantity of 2 instead of 20.
- Allocation of Users: Make sure that every module has an owner, stakeholders, and Make sure each individual has special screens for their interface. Correct user allocation leads to checks and balances as well as focusing on ownership of data.
We have all heard of this phenomenon, optimizing a system, but what is it really about. Well in the supply chain it’s about creating an effective database management system that process data to speed up the supply chain as well as reduce cost in production, and distribution. An optimized supply chain system will enable users to input every bit of data and create formatted presentations, electronic signatures, data sharing, tracking and providing decision making analytics. It will be enhanced with automated hardware such as RFID and barcode readers, conveyors and counting machines. IN the more complex areas, it will even interface with production machinery, receiving as fabrication data directly from one machine to the system, providing up to the second information. (mainly found in pharmaceuticals).
An optimized system does not need to be expensive, or even a full system. The IT you need must match your requirements and the future requirements for expansion. As such, small operations might not need a fully automated ERP/MRP system; they might only need Excel with MS Access system that is bespoke to their needs.
A word about Blockchain
This is not the holy grail of supply chain, it is a useful platform that enables multiple users (nodes) to perform operations (transaction) at the same time, and then close the transaction with a timestamp, which is validated by all the nodes. A blockchain is secure since every one that connects to it should have a copy of the blockchain downloaded on their PC. (this is not the case all the time if you link via a dApp.) The larger the blockchain, the slower and more expensive it becomes to run since each block requires validation, and there are consensus protocols in place to validate transactions.
The current issue in blockchain technology is to create a viable consensus protocol that retains the security of the platform without affecting the integrity of the operations. At the moment, open source clouds are still outperforming any and every blockchain around.
The Bottom Line
Supply Chain IT is like any other IT system. The only difference is that it processes to supply chain data. For a perfect solution consider these basic four questions:
- What is the size of your operation?
- What is your annual P/L?
- What percentage of expenditure savings and improved performance will this system provide?
- How long will it take to install, integrate and teach your staff, before you see the initial ROI?
By answering these four questions, you will be able to decide what budget you have for what system.