Knowledge is predictive power23 May 2016
Laundry consultant Steen Sogaard examines how managers can ensure the right information is available to decision makers
With the right information we are able to create better results. It is pretty obvious, really, that in an industry such as ours where the hourly flow of goods is controlled by operation managers, we create better results through smarter decisions with better knowledge. However, better knowledge is not just more information.
Ultimately, the major part of a laundry's profitability comes down to a very limited number of decisions including call off sequences, process route choices, when to open and close workstations and who to allocate to, where and when. These decisions are made repeatedly, by a small number of people, and impact on 60-75% of the cost complex in a laundry, ie, total variable costs (TVC).
In some laundries it is possible to tell by the results of these decisions who was the operations manager in charge on a given day. In the majority of laundries this is the extent of management influence on results, because apart from a laundry's TVC caused by independent factors such as hourly wages, water levels, bath temperatures, number of pump strokes, programme times, number of feeding stations, and so on, some variable costs are "dependent", or "double variable".
Dependent variable costs are consequences of the hourly decisions about flow and allocation, such as bath exchanges, empty compartments, idling, bottle necks, capacity jams, micro pauses, sub-manning, sub-norm performance, production of rejectable items, underloading, production of unwarranted goods - the list goes on. These high impact decisions are in the hands of the operation managers.
In practice, when giving our managers the task of deciding, we place them in a "decision wrench" made up of, on one hand, the customers' demand for supplies and, on the other, the quality of information available to managers.
If we just give them plenty of textile stocks, the task of making the decisions is made easier and indeed numerous laundries handle the problem in this way.
However, textile stockpiles come at a cost. Literally. Although It is not exactly wrong to pile up textiles in laundries and with customers, and it does help us out of a lot a trouble, it is certainly not spot on the money either.
The right way to handle the problem is:
To adjust the flow of goods in the laundry to the specific market demand;
Adjust resource measures to the flow of goods, and;
Execute call off sequences and employee allocations by means of an operation strategy that aims at maximising the average flow of goods.
It may sound easy, but with machine breakdowns, employees calling in sick, dynamic bottle necks moving around, long-closed process lines, delay between decision and impact, "double depending" variable costs, and so on, it just isn't. The job of an operation manager in a large laundry is not always an enviable one.
What does single out the best managers - apart from being able to handle pressure - is their ability to make the right decisions based on the available information, no matter how incomplete and shaky this may be. He or she will have a keen eye for what to look for, and - somehow - are able to make decisions about the future, based on information about the present and the past. The less skilled ones just do what they have always done, which is actually the right thing to do when everything is as it has always been. Which it isn't always..
Data automation in laundries has developed from first generation with a focus entirely on momentary data, used for example by Programmable Logic Controllers (PLC) to control valves, pneumatics, engines, and so on, through the second generation with focus on data feedback about historical situations.
The third generation data automation is going to be focused on feed-forward, assisting our employees and managers with valuable information in making the right decisions.
It is not the generation of copious amounts of information that makes the difference between "trivial" and "valuable" knowledge. It is the ability to focus on the specifics, right now, among the background noise of trivial data feed from PLCs all over the laundry. It is the ability to wash, process, relate, forecast, and present these specifics ready-to-act-upon, to the right people at the right time.
For example, it is within technological reach to get a signal from each PLC every time it executes an operation, to gather those signals in a database in order to learn how many pieces each operator produced in each category on every single work station yesterday, along with information about stops, jams, failures, down time, maintenance, pauses and so on. However, there is a major difference between this flood of data and a focus on the few specific employees with diverging work station attendance and productivity.
If you calculate the ratio between an employee's actual work speed and the work speed standard, you get a percentage. A good employee's work speed ratio for a whole day may be 106%. If you also compute the ratio between the time this employee spent at workstations and the same employee's available time in the laundry, you get another percentage, hopefully close to 100%, but it could be, for example, 89%. If you multiply these two ratios, for each employee on a given day or week, you get a rank. Some employees' rank index will be around 100%, maybe higher with others around 75%. Now you know who to reward and learn from and who needs special focus on training.
There are usually very good reasons for poor performance, including poor maintenance, bad instruction or just unrealistic expectations and sometimes we, as managers, miss the underlying reasons. Sometimes high performers enliven us with novel ways of thinking and working, ways we missed ourselves. In this way you may learn about what I have come to call "whips" - those with a natural, positive influence on their colleagues - and you may want to figure out where to place them in the laundry to get the maximum benefit from their presence.
If you calculate one site's average rank and compare it to the average rank of other laundries in a group, you can tell which laundry is the highest performer and which laundries and people require closer attention. In the right hands, this knowledge is dynamite - specific, ready-to-act-upon knowledge, pertaining to the biggest single cost in the laundry, wages, making it possible to improve productivity by 10%, 20%, and sometimes even as much as 30%. In this respect, smarter decisions certainly do produce better results.
Data capture is evolving from an add-on to PLCs and machines and is expected to be free of charge just like the PLC programs themselves and primarily aimed at documenting up-times on the machines, into a value creating business area in its own right, thereby adding value by turning trivial information into valuable knowledge, and presenting it in an easy-to-understand and ready-to-act-upon manner.
If we look at comparable development in the automotive, electronics and food production industries, we see that hardware and software are more and more dependent on each other, and that software choice and supply offered is steadily increasing. At the same time software is becoming a separate solution, valuable in its own right, available from independent vendors.
We know that running a business is a constant process of change - whether it is in operations, marketing, finance, or HR - because what our business is about is markets, demands, competition, technologies. All of which are in a constant state of change. In this respect Darwin was right when he wrote: "It is not the strongest that survives, nor the most intelligent - but the one most responsive to change."
We also know that laundries want better information in order to make better informed decisions in every part of their business. They want predictive power. They want visualisations, data breakdown, drill depth, data consolidation, and so on. We need, in the industry, to create and facilitate the link between data exploration, data experience, cognitive experience, predictive power and actions. And we need this link brought to the fingertips of our managers, and the operation managers on the shop floor, without them having to be skilled in databases, scripts, Boolean events, statistics, and so on, and without having to make an application to an IT department in another county.
We get better results through smarter decisions based on better information, turned into predictive power and targeted actions. And better information is not a sample or an average.
Knowing the average car speed between London and Liverpool is 53 mph is not something we can act upon.
Knowing the average productivity in a laundry is 34,8 kg/hr is not enough. We need the specifics. We need to be able to analyse data along timelines, in their full extent, even when they are not linear, to find the correlations and causations, and to work out proper actions. Most of all, we need to be able to single out the right one, and, with some acceptable level of certainty, to be able to predict its outcome.
But, we also know that the current generation of tools does not give us that. The next step of evolution is not Business Intelligence. BI is already available in the industry, either as self-built in-house solutions or standard solutions from outsourced suppliers.
The next step is dynamic, feedforward systems - giving the right people predictive power at the right moment - information solutions supporting our managers in making the right decisions when planning the coming hours of production, such as expert simulation, best practice, manual or automated optimisation systems. We have already seen the first glimpses of dynamic optimisation solutions in the industry.
Beyond that, we need the software solutions to merge and consolidate into "turnkey packages", as we have seen happen with our laundry hardware. We know from experience of that, the potential of integration is large.
The interesting thing is to see who is going to rise to the occasion - one of the existing suppliers of machines, chemicals, RFIDs, textiles or software? or somebody with the right technology, but from a kindred industry, on the look out for new