Software estimation and outsourcing complexities in manufacturing solutions

Complexity #1: Gathering detailed user requirements

Whenever any software development project is started, it starts with an objective which has some defined requirements and functionalities underlying it. One simple example is – development of Industry 4.0 point solution for a manufacturing industry. A complex example could be – development of Manufacturing Execution System (MES) for a manufacturing industry. So in both examples, we have a clear objective to achieve but if we go bit deeper into this objective then we will find more detailed requirements. In case of the point solution example, besides technical stuff of deployment / hosting and domain etc., we need to define various pages, sections, contents, images and reports etc. On the other hand, in case of complex MES system development, more complexities are involved in terms of various stakeholders, technical and functional details, machine connectivity needs, types of securities and system integration needs etc.

To build a customized software, the requirements are generally never clear prior to requirement gathering phase. Requirements keep evolving. Even if, they are well-documented in RFP (Request for Proposal), they are never synchronously understood by all the stakeholders. Additionally, because of the new on-premise web or cloud platform, there will arise various non-functional requirements like performance, multiple browsers, multiple devices, dependency on other IT-OT system, data security, network bandwidth etc. which also invite complexity in the project. When you are quoting during software sales, your estimate can go off by more than 50 %.

Complexity #2: Unclear requirements

Whether the project is small or big, every project with some objective will have huge set of clear and unclear requirements. These requirements may be clear at the beginning of the project or will be clear as we progress on the project development. The uncertainties about the complete requirement at the beginning of the project makes the project estimation a complex process. Sometimes an unclear objective will also add up to complexity in estimation!!! Another biggest issue is the uncertainty involved at the beginning of the project. Many times even the customer is not clear about the whole requirement of the solution needed. If there is no complete clear requirement then how it’s possible to estimate it in term of effort and time?

Complexity #3: Integration of new solution with existing IT-OT systems

Many a times in complex projects, system integration task plays a key role. But software companies sometimes underestimate efforts for system integration tasks. Explaining complexities involved in system integration to a customer is tricky. They may say that you can just take the API data and push it through. Yes, you could. But you then have to test it and make sure everything works together. And then, when you throw in multiple devices and platforms – like web applications, backend services, Windows applications, mobile applications, HMIs of machines — all of this suddenly exponentially grows to the point where you turn around and say to a customer that to test all of this properly is going to cost additional thousands of person-hours!!! Read my blog on System Integration Complexities for more details.

Complexity #4: Prediction of reusability factor

This is applicable when roll-out of developed solution is expected to happen in multiple branch offices of customer with some level of customization in the functionality is foreseen for each branch office.

At the time of proposal submission, it is very difficult to predict “% code reusability” factor where the solution is expected to be deployed at multiple branch offices of customer that involves customizations in software functionalities for each branch office. Normally software companies put lot of efforts while developing a “product” where “generalization” of functionality becomes very important which becomes “base layer” of the product. However software companies will not get sufficient time & cost to do the same in “project” development.

Complexity #5: Choosing right technology stack

These days, technology is moving very fast & continuously changing. There are patches and versions of each technologies which you need to keep an eye on. Patch or new version of technology might have resolved some vulnerability or major issue in the technology stack. By the time you start software development, some new technology or a newer version is beckoning you and it is tough to decide whether you opt for the current stable version and compromise on the new features or try the new version with some amount of risk.

Complexity #6: Outsourcing of work packages sounds easy, but difficult to manage

No one should deny that outsourcing of project can bring several benefits to a company. But no one also should forget about the challenges that we face by outsourcing the work (whole or part of project). Challenges are like – misunderstandings in communication, mistakes regarding technology, misinterpretation of bad quality requirement specification document, incompetent engineers in outsourced team, outsourced team sitting remote location and many more like this. This way challenges will occur at every stage of the outsourcing service – starting with the selection of the vendor, through the implementation development process, up to the delivery of the final product.

The key to success is to choose the right IT supplier and determine the course of collaboration in detail. Also, ensuring high quality requirement specification is very important while outsourcing any part of the project. If you take care of all such aspects seriously, you will not only avoid problems but also receive a high-quality deliverables from them.