< back

10 reasons why so many MES projects fail

August 29, 2013

10 reasons why so many MES projects fail

It is a well-known fact that most IT projects fail, if not entirely, then in at least a few of their deliverables. They might take more time to complete, higher cost than planned or proposed, or just fail to meet the expectations of the company purchasing the software. What most people don’t realize or care to analyze is why such projects fail and how can such failures be avoided.

Generally the main reason for enterprise wide applications to fail is not even related to technology, but instead to the management of the whole endeavor and involvement of business leadership in the design and implementation of the application. IT applications fail when they are considered as separate from business and its underlying goals. A good McKinsey article on why global IT applications fail to deliver can be found here.

When we consider an MES application, the chances of failure may be even higher simply because of the nature of manufacturing and its inherent complexities. These applications are expected to address every aspect of the manufacturing operation, which constitutes a complex web of components which broadly classified are- material, man and machine. The software should be robust and at the same time flexible to deal with contingent situations, process changes and new technology.

Every manufacturing process is distinct from the other in some way, even if it looks similar. Data is continuously generated and every event, no matter how trivial it might appear, has an impact on the overall operation. An application like the MES needs to integrate the process, with its components and then provide real-time, accurate and actionable information from the process, and this is easier said than done.

So the organization, project and technical complexities lead to the following main reasons for failure of MES projects:

1. Project must be initiated by top management and aligned with business goals. It must have management’s full commitment throughout the project. Generally what leads to project failure is the absence of management’s active participation in review and progress of the application and the functionality it delivers. Frequent steering meetings in which progress is reported are good practices.

2. Expectations need to be realistic. Both in terms of the project plan and in the improvement results. When we consider a complex and multi-faceted application such as the MES, it is crucial to understand that high expectations may lead to both failure and frustration. Ideally, organizations should undertake a pilot project for a particular segment of the production line/operation, then based on the actual improvement define realistic expectations for the entire plant. Such a modular approach can prevent sunk costs, rework, save time and allow the management to experience the level of improvement attainable after a full-fledged implementation.

3. Production departments should drive the project. Due to the technical nature of the software product, many times the project is driven by IT. However, when this happens, the misalignment with the business goals and processes is more likely. It is extremely important to realize that an implementation of MES is a project of the entire organization, which should be driven by manufacturing, where IT has obviously a very important role. In any case, a neutral project leadership is a good idea, preferably from the top tier of management would ensure that the interests of all stake holders are heeded and used, with the aim at all times being to make the operation more efficient, effective and indeed more profitable.

4. Processes must be mature. If the process is still immature and not optimized, it is important to realize that an MES system may do nothing else than automate existing inefficiencies. Processes which are in nascent stages generally have a lot of areas where improvement can be achieved by simpler techniques using the WBS-Work Breakdown Structure or ABC-Activity Based Costing or Kaizen or any other improvement techniques depending on the process. Only when the process has reached the know-how level in the learning curve there can be a possibility of successfully implementing the MES, to reach the Know-why stage. The project is bound to fail, no matter how impressive it’s features, if the process owners are unaware of their process in its operational entirety.

5. The right MES must be selected. Organizations which buy MES applications off the shelf, and try to apply them to their process with minimum customization, often fail to achieve the desired results. They are the victims of the- ‘One size fits all philosophy’. Applications like the MES need to model the process and provide ways to improve it. When the application is made to fit the process, the probability of improvement increases, simply because the process will remain as is, avoiding resistance from the work-force and also helping to achieve the desired level of performance.

6. The right MES implementation company must be chosen. A huge and almost irreversible mistake that organizations commit while considering an MES implementation is choosing a system integration company which has no experience in the MES space. What organizations need to understand is MES is a complex application, which will encompass the entire operation. Experience and expertise in MES implementation are highly recommended as this will not only save time and cost, but help the organization achieve a solution which is accepted by the users, provides better functionality and makes the process more efficient.

7. The existing IT landscape needs to be considered. Even though there might be no other MES earlier installed, except in the case of a completely new factory, it is very unlikely that there are no other applications on the shop- floor. An MES is supposed to integrate the operation along with the existing IT infrastructure, so the application would be a failure if separate systems exist and users need to now work with both these and the MES separately. MES application needs to be a single comprehensive platform for optimum results. This means either legacy/older applications should to be integrated to it or simply replaced. The difficulty is the time and effort this analysis and subsequent implementation take.

8. The system must be tested, go-live rehearsed and project re-planned from the learnings. Before going live if the team fails to test/rehearse the system, learn from the issues, re-plan and re-implement, the MES would have a high chance of failure. The idea of testing the system is to ensure that it will perform at-least above the minimum satisfactory level. When problems are detected after go-live, the disruption caused by trying to change/improve the situation can prove very costly and time consuming.

9. All users must be trained. They’ll need to be fully aware of the changes and might provide very valuable input, still in time to be incorporated. Almost every software implementation demands changes in the ways things are done, which clearly mandates that the people, who will be affected, should be made aware of the change and how to deal with it. If this is not done, it could lead to high resistance and even complete failure. It is simply unforgivable not to pre-train and involve process owners in every step of the MES implementation. Training will help work-force warm up-to the application and help improve the solution, as no one knows the process better than the people who drive it every day.

10. The project doesn’t end with the go-live. Actually, it is almost the start of a continuous improvement process. TQM theorists claim that ‘perfect never is’, i.e. no matter how efficient a process appears there is always room for improvement. This applies even to a successfully inducted MES application. It is the nature of processes to change, which is dependent on many factors, such as the market, material, technology and now of-course more than ever-knowledge. The MES project starts adding value to the process, once there is a successful go-live, after which the challenge is the continual addition of value in-terms of improvement of process knowledge, efficiency and overall effectiveness. To leave an MES after a successful go-live, would be a failure to acknowledge the constant change and thereby a failure to continually improve.

All in all, any IT endeavor can fail when its end-goal is not understood and communicated. MES implementation requires careful planning and support. Its success depends largely on the way it is perceived by the top management and various stakeholders and that it should be made as closely as possible to represent the process, with a reliable and technically adept vendor. However, if driven properly, an MES can prove to be much more than a cost-saving and quality improving tool, it can actually be the vehicle for process excellence.

< back


IIoT Has a "Thing" for MES. Why IoT Platforms Won’t Replace MES for Industry 4.0
by Iyno Advisors and Critical Manufacturing