Learnings from Boeing

This article is sort of relevant to my previous post on 'Project Phenix'. We discussed the commons aspects of manufacturing and software development. I saw this video in youtube. Please take a moment to watch this video.

Remember, they are building aircrafts. There is no rollback option available for them. There is no plan B when are flying. Yet they build eight 777s per month.

Read this article too for more information: http://www.businessinsider.com/how-boeing-produces-777s-2012-10?op=1

So, what we can get from this?

We know the level of quality that they are looking for. No need to discuss about the quality of the product which flies million feet up on nowhere in the sea! alone!.

Looks at assembling! very similar to each unit in software development. Each module is independently testable and can be easily fixed into the system.

• Independently testable modules- each module has undergone a series of quality checks before it actually fixed into the main system. It dramatically reduce the likelihood of failing the unit itself. They simply fix the unit into the plain. Can we do it in our projects? software industry in general is far behind the manufacturing in this context. May be for valid reasons but we are still behind. We do not think of quality in the unit level. We do not think of the test coverage of the unit. In fact we do not have a framework to test in the unit level. We do not have a framework which makes sure that the quality is in built in the unit level. We do not have a framework which assures the compliance of the code for monitoring and the compatibility with test automation framework.
• Under one framework, deliver a flying machine - each unit they produce, each nut and bolt they buy from suppliers, each manual they compile, everything is compliant to one common framework of quality. I am sure that each of us have seen, good teams who deliver fast with quality, surviving team, struggling teams. Are we under the same quality framework? if so why team A has more production issues that Team B. Its because we are not under one quality framework. The guy who fix the engine to the aircraft does not have to bother on its quality. The guy who comes to fix the seats knows that the holes are grilled in the right place and its just a matter of putting the nut and the bolt. Only few companies like Google have fixed this problem. All of their products come under compliance of one quality framework.
• Automation - see how much of automation they have used. Automated drilling has brought tremendous results in terms of saving rework and time of laborers. They have used automation as much as possible. Note that its not 100% automated. They have not invested billions of dollars to fully automate the system yet they have automated routine and time consuming actions of the process. In software development, we might not need to automate everything. In fact its a utter cost. We should automate the stuffs which brings value and saves time. Deployments, Test Automation, Monitoring are some of the areas we can save time.
• Less work in progress in each work station -did you see the plain stops in one work station? No its just moving. There is very less work in progress in each work station. No pending work filing up in the work stations. Optimization of workstation is one of the key things we can learn from manufacturing. 
• Continuous focus on improvement: Its a journey. It never ends. Focus on improvement is the key aspect. If you read the article which I shared the link above. You can realize how wiring section speed up their work by just grouping each wire sections. It improves accuracy and time. The quality is in built. No no one doesn't need to double check whether the wires are grouped properly or damaged during the separation.
• Availability of tools - just imagine a laborer is waiting for another laborer to finish his work to get his screw driver. Its crap. Similarly how many times we have wasted time to fix environment issues? how many times we spent late evenings to cover up work? This is a common issue in the entire software development forum. Note that the laborer does not have two screw drivers. Point is not to waste or invest money to increase the numbers of existing tools but to new tools and techniques.