From Legacy ERP to Smart Industry 4.0 - The Pace-Layered Application Strategy Model

  • April 11, 2016
  • Feature

By Jan Baan, Chairman, Vanenburg Software

According to Gartner, companies should adopt the Pace-Layered Application Strategy to build an application portfolio that supports new business ideas and the inevitable business changes.

The Pace-Layered Application Strategy Model from Gartner distinguishes three layers, which are helping organizations improve their applications by categorizing, selecting, managing and educating themselves on those applications.  As a result, organizations are finding that it is more efficient, faster and more economical to change, differentiate and innovate, as well as realize returns on their investments. 

Systems of Record

In the 1980s, relational database technology and MRP algorithms helped us further expand our logistics systems to so-called Enterprise Resource Planning (ERP) systems. In addition to planning and stocks, the financial data were linked, thereby simplifying periodic reporting to shareholders. Therefore, the ERP solution belonged to the core systems of the company, which also has, unfortunately, led many users to feel locked into their software vendor since it is the software vendor that programmed those core algorithms. 

In addition to the aforementioned techniques and the UNIX-based client/server infrastructure, the event-driven 4GL language was critical to building these complex systems. In the 1990s, SAP and Baan Company (purchased by Infor) were the market leaders in complex ERP systems. These systems functioned as an “on-premise” solution and were developed as single company solutions at a time when the Internet was not used for business.

This was a good and profitable business for traditional software companies because the current maintenance contracts became the basis of their long term profitable revenues. However, innovation was unfortunately not required to maintain these revenue streams. How can we explain this lack of innovation?

At the end of the last century, we faced Y2K and all innovative capacity was used to fix that problem in already-outdated systems and make them suitable for this century. This operation was successful and these legacy systems are still serving us after January 1, 2000. The consequences of this significant development effort to convert these legacy systems to support the dating of 2000 (and beyond) didn’t leave budget money for new development and, therefore, all innovation came to a stop.

At the beginning of this century, Microsoft acquired Axapta, Navision and Great Plains and was forced to combine these systems from the 1980s, especially for smaller companies, in a project that cost billions: Microsoft Dynamics. With the Microsoft .NET Framework for Windows with .NET Framework Class Library (FCL), the necessary "platform or platforms" was not developed. Microsoft Dynamics was no substitute for the complex ERP systems such as SAP, Baan, J.D. Edwards and Oracle.

Needed - A Minimum 4 Types of Systems.

It should be noted that manufacturing businesses require at least four types of systems to provide value to their enterprise.  In addition to ERP, companies find CRM (Customer Relationship Management), PLM (Product Lifecycle Management) and SCM (Supply Chain Management) to be indispensable to their business. All of these systems originated in the last century before the Internet was active. They are characterized by their complex database structures of 10,000+ tables. For example, SAP uses 60,000 database tables.

The need for these systems led to the extreme growth of System Integrators, or SIs, at the beginning of this century. Through their handicraft approach a lot of customization was added to these complex systems and the legacy applications became even more complex. Subsequently, the cost of all that customization as well as the implementation of the solution became much more expensive than the cost of standard software. For at least ten years, though, it has been technologically feasible to disconnect this standard solution as a vanilla solution and enrich the legacy systems with the help of the newer technology in the process layer (Service Oriented Architecture or SOA).

While the System Integrators functioned in this traditional approach, it was not good for customers. A good example of this phenomenon is the Speer project at the Dutch Ministry of Defense, where the traditional customizations performed by SI’s in a highly complex SAP system cost the customer more than one billion euros. Despite these huge sums of money and the attendant frustrations, it was determined at the conclusion of the project that this complex arrangement was not workable.

Because of this traditional course of action, the SIs also ensured that the old legacy systems appeared to be irreplaceable, which has allowed the major software vendors over the past 15 years to simply lean back, ignore any efforts at innovation and collect their maintenance fees. The SIs could continue their profitable traditional business.

Systems of Differentiation

Early this century, a new idea was introduced – integrate these systems using a Service Oriented Architecture (SOA). Using standard accepted BPM (Business Process Modeling) notation made it possible to assemble many individual silos (composing) as a business management End-to-End (E2E) process and to offer this as a cloud solution for every participant worldwide. With our company Cordys, we took the lead with these BPM solutions and we have developed innovative solutions such as an AOG (Airoplane on the Ground) solution for Embraer and a Spare parts solution for Siemens, which sped up the maintenance.

Characteristic of these BPM solutions is the ability to build new solutions based on case management processes next to the transactional systems that were integrated from these silos. This enriched the uncompromising transactional backend solutions by using the new technological possibilities of a case-driven business process. These processes could not be controlled with the transactional systems, so spreadsheets were introduced as a tool for the knowledge worker. Microsoft Excel is often called the most dangerous software program due to the lack of manageability and compliance. Now, with dynamic case management, people can manage the business until close to the delivery to the customer.

In 2009, Embraer started to implement significant process and organizational changes designed to improve efficiency, provide greater transparency for management decision-making and enable service improvements in key areas. Soon, it became clear that each proposed change would require significant engineering work at the core SAP level and this would be time consuming and costly. To leverage their investment in ERP while increasing flexibility and agility, Embraer selected the ERP ShapeShifter solution from CSC, built on top of the Cordys Business Operations Platform (BOP). Cordys helped Embraer to achieve business efficiency improvements while still successfully managing the core underlying technology issues.

Embraer was trying to achieve an AOG (Airoplane on the Ground) solution with SAP ERP and Siebel CRM. This was difficult to realize because of the rigid structure of the transactional systems; the AOG business process is case driven. The Cordys application integrated with SAP and replaced the Siebel solution. The Cordys solution has led to improved quality with consistent, standardized processes across multiple systems and an enhanced user experience. Embraer now has real-time proactive monitoring of vendors’ quality index that helps them to reduce potential cost and delays in the assembly.

Embraer had enough confidence after implementing the Cordys platform to discontinue the SAP maintenance contract that had cost them millions. The innovative BPM solutions were particularly popular at insurance companies given the significant number of rules that are inherent in their forms and business processes.  

In production companies, however, BPM solutions were not as evident. CIO’s usually did not recognize that there were innovative ways to isolate legacy systems, such as a vanilla solution without customizations in the 1st layer: Systems of Record, and, then, to decouple and adapt this to the internal business process in the 2nd layer: Systems of Differentiation. Despite the fact that these inbound business processes within an enterprise’s BPM Cloud solution were much more manageable, the reduction remained limited to structural data. There was no meaningful use of the consumerization of IT made possible by the smartphone focused on the Internet of Things (IoT) and big data.

Meanwhile, over the last few years, we have seen the growing popularity of SaaS (Software-as-a-Service). It is in this market that Salesforce managed to oust the previously dominant Siebel with their innovative CRM system. These experiences have made Salesforce into a market leader in the PaaS (Platform-as-a-Service) domain with its Salesforce1 PaaS platform, for which there are now thousands of PaaS solutions available.

The inheritable objects in the Salesforce1 platform makes it easy for developers to enrich these objects with their own Intellectual Property (IP), where database queries are greatly simplified and it is possible to model their own business processes. This was not possible in the transactional systems from the 1st layer. Instead of 10,000+ tables, we now only have to manage a few hundred objects.

Consider an automobile example. You could compare these two different techniques using a 1990 Jaguar and a 2016 Tesla. Both products have mechanical and electronic components, but Tesla adds software components. The car is no longer delivered as a final solution from the factory but can be continually improved through innovative software solutions. The Jaguar components were built in the last century and are therefore obsolete, but the real breakthrough was the reduction of the number of components. The Tesla has only 15% of the number of components of the aging and fuel-guzzling Jaguar but runs on clean and inexpensive electricity. We can imagine that if you like driving classic cars, the Jaguar will be nostalgic and enjoyable to drive, but you would not want to rely on it for daily driving.

The elasticity of these available industrial solutions systems not only reduces the cost of implementation rollouts by a large amount, but more importantly, the systems remain up to date because they are renewed regularly without any negative effects on the user.

The innovative Cloud ERP system of Rootstock Software on the Salesforce1 platform has been available for the past few years. Rootstock management and staff has extensive experience in manufacturing and distribution over the last 40 years, and this experience can be seen in their solution’s broad and deep functionality and flexible architecture. Salesforce is one of the shareholders in Rootstock. Salesforce1 allows companies to integrate their value chain. Currently CRM, ERP and SCM apps can already be integrated on this platform, and there are thousands of other apps available on the platform as well. In addition, all-important Industry 4.0 functionality is supported directly.

Industry 4.0 - Digital Enterprise

Smart Industry 4.0 is the Fourth Industrial Revolution. The IoT is often mentioned as the fourth industrial and economic revolution. The four revolutions are Economy 1.0: Steam engine, Economy 2.0: Mass Production; Economy 3.0: Digital Revolution/Internet Technology; and Economy 4.0: Mobile Internet and Internet of Things.

We are talking now about the Digital Enterprise. But this is more than a translation of the structured data of output documents to a PDF file.

From the beginning of our third startup at Vanenburg Software, we have been working on smart process apps. We see the document as a dynamic shape - a type of XML file. In this 3rd layer (Systems of Innovation), are the business processes, which are much simpler than in the 1st layer (Systems of Record) and 2nd layer (Systems of Differentiation), where the data is still stored in structured databases. The much simpler workflow processes in the 3rd layer inherit the modern objects mentioned in the 2nd layer, or can be integrated with legacy systems from the 1st layer.

In the 3rd layer, one can see the innovative app as a disruptive business management process. With our experience of case management in our second startup Cordys, we succeeded to add the case-driven workflow to the dynamic file. This can be seen as a kind of Business WhatsApp.

Our product Collabrr facilitates the modern knowledge worker to cooperate in a network to both produce secure documents and easily create contacts for the maintenance and sharing of documents such as texts, spreadsheets, presentations, files and UI screens, with business partners.

The emphasis will be on the productivity of the knowledge worker who collaborates with his supply chain colleagues. Apps such as Procure-to-Pay are ideal for document-driven business processes apps. These apps can be enhanced with information from the IoT (Internet of Things) and Big Data.

The procure-to-pay systems enable the integration of purchasing department with account payables department. Some of the largest players of the software industry such as Oracle, SAP/ARIBA, B-PACK, Procurify agree on a common definition of procure-to-pay, linking the procurement process and financial department.

This on a state machine-based workflow is “event-driven”.  On condition-based transitions can be inserted into different states or steps, such as events or notifications. As a result, powerful apps can be built that may be integrated with, a complex external rule engine, for example, or with complex machine- to-machine communication.

With the Internet of Things, we are able to connect much easier to new devices. We see interesting initiatives like Atomiton's TQL System which provides "Supervisory Execution," just like a human nervous system for interconnected things: sense, respond, regulate, and execute.   The Thing Query Language for Thing applications (TQL) is to things such as SQL is to data.  It is an IoT platform that allows you to manage huge numbers of IoT devices which can communicate with each other, sort of like a Facebook for things but safe.

It lets people create real-time context responsive applications by scaling and streaming from numbers of sensors and control devices. It goes beyond data. For instance, with a simple query - Go to all rooms with >35°C temperature and update their fan speeds to 850 RPM - users will execute real actions, not just create data, immediately.

Learn More

Did you enjoy this great article?

Check out our free e-newsletters to read more great articles..