Today there is a new concern for managers in many manufacturing companies: having a rebel department doing everything out of control.
Maybe you are wondering what exactly means “out of control”? Good question. When the products of your process are generating a register outside the control limits, the process is out of control. In picture below, the circled point means nonconforming results of process.
(Edited from original, http://pinterest.com/pin/401453754256413495/)
Statistical process control is a powerful collection of problem-solving tools useful in achieving process stability and improving process capability through the reduction of variability. A process is said to be in statistical control when it operates under only chance causes of variation (Taylor, 2009).
All-over the supply chain you can apply statistical control, the scope of this tool includes early detection of issues which could unchain disruptions, shortages, overstock, expensive transportation, bad suppliers and other risks in logistics. Besides it is an accurately way to measure and compare the performance through time and basically it is a cost reduction trigger. We talked about measuring in past columns; it´s time to bring it to practice.
To implement a statistical control process, the first and most important step is defining the process that you want to control. The better defined that your goals are better the results. You can work in a cost deployment, to identify cost reduction opportunities; with this tool you will determine where is leaking the process and set your goals clearly. (Szwejczewski & Malcolm, 2013)
Then we need data, the data that you already have is the beginning, but is important to wonder if you need more information in order to get more accuracy in results. Concurrently, collect the voice of customer and the voice of process, which will help you to delimit the tolerance of process (Cudney & Kestle, 2010).
Now you are ready to calculate the statistics. Choose carefully the chart that fits better your process and establish your control limits. The formulas can be found in any statistics book; remove dust in the bookshelf.
Let´s understand that standard processes have standard outcomes; when you add a variable you will start to see different results.
A good indicator for transportation is the cost of freight as a percentage of sales; I think this is the best way to measure freight because it gives an evaluated number per se. The 23rd annual study of Logistics and Transportation trends reveals that 73% of companies responding the study maintain the percentage of freight cost between 1 and 5 per cent of their sales (Holcomb & Manrodt, 2014), which is a healthy number depending of industry and size of firm. So, controlling transportation charts will let you know if you are part of the elite of world-class manufacturers.
The chart below, which is fictitious for illustration purposes, shows a good example of evolution from what I call “wild logistics” to controlled logistics.
Please note that 2013 and first half of 2014 have variations between 4 and 7%, some months were good, others without comments. Red points are variables out of standard, such as late deliveries, a different carrier, premium freight, a rate change, etc. So, the problem is identified and accepted, what is next?
The control limits were set in a range between 3.3 and 4.3. Each red circle needs an action from your organization to avoid that those variables occurs again, the goal is reduce the variation of process to guarantee standard outcomes and finally have control of your department. You will see numbers fitting the parameters that you determined; voilà wild logistics were tamed.
Inventory reduction is a “lean” basic. Actually, inventory is identified as one of the seven wastes of Taiichi Ohno, it is perhaps the most visible form of waste. It represents between 5 and 30 percent of the total assets of a manufacturing company, depending of the industry could be bigger like retailers. (Goldsby & Martichenko, 2005). Statistical control can be applied as well to help you to reach your goals of inventory. Let´s imagine the situation: a point under the limit of control means shortage and upper is overstock. You can measure it in days of inventory or value; in all cases, if you are able to set correct control limits you will detect discrepancies –variables– before chaos.
Remember, a control chart only describes how a process is behaving, not how it should behave (Taylor, 2009). Your actions will drive you to control the chart; therefore I recommend record each action and its effect in your performance. A good combination of measuring and action plans is the way to success. Good luck with the beast!
Thanks for reading and share this article. Until next one!
Alan Rodríguez. @AlanRodEsp
Cudney, E. A., & Kestle, R. (2010). Implementing Lean Six Sigma throughtout the supply chain. Goldsby, T. J., & Martichenko, R. (2005). Lean Six Sigma Logistics: Strategic development to operational success. Florida: J. Ross Publishing Series. Holcomb, M. C., & Manrodt, K. (2014, Sep 1). Logistics Management. Retrieved Jul 20, 2015, from 23rd Annual Study of Logistics and Transportation Trends: The transportation tug of war: http://www.logisticsmgmt.com/article/23rd_annual_study_of_logistics_and_transportation_trends_the_transportation Szwejczewski, M., & Malcolm, J. (2013). Learning from Worls Class Manufacturers. London: Palgrave MacMillan. Taylor, G. D. (2009). Introduction to Logistics engineering. Boca Raton, FL: CRC Press Taylor & Francis Group.