Parent-child relation from the bill of materials
Sometimes you may experience that even though everybody in the production are working hard and productivity is high, you can’t seem to get the trucks going.
Nothing seems to be completed until everything has been pushed through the production. In manufacturing industries this effect is caused by bills of materials. In our industry the effect is the same, but caused by and called something else.
We know it from the manufacturing industries. A chair can only be assembled when the back, seat, legs, screws, bolts, nuts and everything are finished in the production and are ready for assembly. It is this parent-child relation from the bills of materials, and its need for synchronization, that limits the assembly.
In the laundries the effect is also caused by parentage, not on the product level though, but one and two levels higher – on customer and route levels – and is called requisition or route sync effects.
Route sync effects
The articles from a customer are sorted in the check-in and sent into the production. Only when all the articles from each customer are processed and gathered again in the dispatch department (according to requisitions and route pick lists) we can get the trucks going. And only then.
This way a loaded truck might wait – sometimes for half or full hours – for a few pieces only. As a result not only the batch sequences into the production, but also the sequences on each workstation, become important to truck departures. And both are governed by organisation. With a multitude of batches from different customers in the buffers, sequences become important to the amount of work in progress and to the distribution.
If we leave it to the operators to choose sequences, understandably, they choose the sequences that best fit their own purposes. And we cannot blame them. But by doing so they do not necessarily take the laundry’s needs into consideration.
With many batches, categories, workstations and customers, the batch flow is in all probability not synchronized and coordinated to suit the laundry’s purposes. Then one of two things happens: Either the trucks are not able to leave in time, or the stress level in the laundry increases considerably. At worst, nothing is finished until everything is finished.
So batch flow control also means control of sequences on each workstation. Fortunately we produce in pools (make to stock), so we do not have any problems with late deliveries and stress in the production – some laundries are likely to say.
Really?
The problems do not disappear in pool productions. You may have drowned them in overloaded buffers and stockpiles. Tied up the working capital. But the job is to limit and control money deployment, which is both total variable costs and working capital tie-up. Not either … or. There is always a need to synchronize and coordinate, also in pool productions. And reductions in costs must never happen at the expense of tying up working capital.
It may sound dictatorial, as if only one or a few persons are to decide everything in the production, but it all depends on how, when, and to whom the orders are given, and what information is provided along side. Otherwise even a train departure may appear dictatorial.
But when we, the passengers, are given the proper information, in time, concerning departures, destinations, travel time, etc., the dictatorial element disappears. Then we are equal to the situation, and are able to plan ahead, take measures and make decisions concerning our own work.
Anxiety and stress
And finally organization intends to create an atmosphere of security and confidence. Regulate insecurity. In a chaotic world one of the ways to confine uncertainty is by making hierarchies. And psychologists can tell us that anxiety and stress are best reduced by:
• classifying people in small groups (4-5 members)
• of different ages and both sexes
• in good physical form
• where group members are well acquainted
• with a reputable, fair and generally accepted leader
• in a light, dry, silent, temperate environment
• where everybody is in sight of each other
• not isolated from the outside world
• with a well defined, common target
• and time limits well within reach.
Then we also know how to stress and frighten people, namely by doing the exact opposite, for instance:
• isolate a person
• out of shape
• among groups of strangers of same age and sex
• in a hot, noisy, damp and dark room
• isolated from the outside world
• with individual, shifting, and vaguely defined tasks
• and under time limits way out of reach.
One hardly dare ask: Have we done everything we can to avoid stress and anxiety in our own laundry?
Manning
The variations and cost distribution taken into consideration, one of the most important planning areas in the laundry is the manning of the production – especially in and out of seasons. Two methods for workforce planning single out. Both are rough in the sense that they do not include the production planning’s influence on productivity.
Even though we now know that batch sequences and process route choices exert great influence on the emergence of bottlenecks and on general productivity, it is not possible to swear in the production planning when we carry out the calculations manually. To that we would need raw computing power and an optimization system.
The elaborate
For the laundry with fluctuating demands (categories and volumes), the category-based method is to prefer because it involves every category and every workstation. In short you work out the volumes expected from each customer in the coming season – category-by-category, and week-by-week. Use last years figures, taking into account the influence from this years holidays, World Championships, Olympic Games, foreign exchange quotations, weather forecasts, and what ever you expect to impact the work, preferences, habits and family life of your customer’s customers.
Sum up all volumes, still category-by-category, and week-by-week. This is the model’s input – the laundry’s weekly product mix. Inside the laundry the workstations have to be manned according to this forecast. Since the method works with volumes per category on the input side, we have to do the same on the processing side.
Productivities have to be worked out category-by-category, and workstation-by-workstation. Use weight as common denominator, e.g. clock process speeds per kilogram. Calculate the necessary time (in clock hours) to produce the week’s volumes of each category, workstation by workstation. Then multiply by the number of operators and sum up the weekly working hours on each workstation. Do not round off numbers yet.
Sum up the necessary working hours from all the workstations, divide by standard weekly working hours and you get the size of that week’s workforce – provided you are able to level out volumes over the entire week, and have access to flexible workers.
Put into a spreadsheet, with customers and their volumes on the input side, workstations and productivities on the process side, and the number of employees per workstation on the output side, you are now able to calculate your way through the entire season, week-by-week.
The quick and dirty
The laundry that experience an almost unchanged product mix over seasons, have no reason to distinguish between categories, and would probably prefer to use the quantity based method. If for instance the total quantity in the laundry rises from 1,000 to 1,300 tonnes, and the rise is evenly distributed on all categories, there is no need to distinguish between different categories.
It is probably not the case for the laundry’s entire product mix, but if variations are relatively small, it is okay to pretend. In short you work out the total quantity of linen and textiles you expect to receive in the laundry in the coming season, week-by-week.
Use last year’s figures as a starting point, and adjust according to your expectations for this year’s season. This is the input. Since the method works with total quantities, you also need to work out productivities in totals, for each workstation. But even though the total product mix does not change from last year to this year, it changes gradually in and out of seasons, which most often is reflected in productivities. The same way this year, as last year.
When clocking process speeds, you are likely to experience an increase in productivity with increasing volumes. So when e.g. the check-in handles 5 tonnes a day with 3 operators, but 8 tonnes a day with 4, we have to bear this in mind when we work out the manning requirements, and assess productivities week-by-week in the transition from low to high season.
Calculate the clock hours necessary to produce each week’s total quantities on each workstation, and convert this figure to the necessary working hours. Sum up working hours from all workstations, divide by the standard weekly hours, and you have the size of that weeks workforce – provided, of course, you are able to level out volumes over the entire week, and have access to flexible workers.
Put into a spreadsheet, with total volumes per week on the input side, workstations and productivities on the process side, and the number of employees per workstation on the output side, you are now able to calculate your way through the entire season, week-by-week.
The manning plan
In both cases the most favourable way to place holidays is the same. Going in to, and out of, a season the size of the workforce may change each week. Turn this circumstance to your advantage.
Place the base staff’s holidays adjacent to the newcomers’ start and termination dates, in order to distribute starts and terminations, and prolong engagements. It is easier to absorb newcomers one at a time, than a whole bunch all at once. In the figure below the vertical Y-axis represents the number of employees and the horizontal X-axis represents the weeks over a season. The darker vertical bars illustrate the total number of employees engaged each week. The lighter horizontal bars illustrate the permanent staff’s holidays.
Where the latest arriving newcomer (in the top of the figure) should only have been hired for 3 weeks, holidays prolong her engagement by 1 + 3 weeks before, and 3 + 1 weeks after, giving her a total of 11 weeks, which is often easier to “sell” than 3 weeks, and surely makes less “noise” in the organization.
Also use holidays to reduce the number of newcomers in each week. Instead of taking on four newcomers in one week, holidays reduce the number to two.
Hans arbejde er baseret på års erfaring med virksomhedsdrift og ‑planlægning på mange niveauer, både i produktions-, service- og udviklingsvirksomheder (bl.a. 8 års erfaring som vaskeridirektør).