http://www.bastiansolutions.com/blog/index.php/2011/12/21/how-a-controls-engineer-does-christmas-lights/

Worth the trip to the Bastian blog site to see this!

Payback Period is the financial metric that tries to answer the question: How long does it take for an investment in a project to pay for itself? Or, how long does it take for my incoming returns to cover my costs? Or, put still another way: How long does it take for the investment to break even?  Payback period refers to the period of time required to “repay” the sum of the original investment. For example, a $1000 investment which returned $500 per year would have a two year payback period. Payback period intuitively measures how long something takes to “pay for itself.” All else being equal, shorter payback periods are preferable to longer payback periods. Payback period is widely used because it is easily understood and, when used to compare similar investments, can be quite useful.

There is no true “formula” to calculate a project’s payback period, except for the simplistic and unrealistic case where the project in-flows are constant (or constantly growing) and easily divisible into the initial investment. In most projects, to calculate the payback period is it necessary to create a spreadsheet that looks at cumulative cash flow to determine the period in which it goes from negative to positive.

All thing being equal (which is rarely the case), the project investment with the shorter payback period is considered the better investment. The shorter payback period is preferred because the investment costs are recovered sooner and are available again for further use. Also, a shorter payback period is viewed as less risky. It is usually assumed that the longer the payback period, the more uncertain are the positive returns. For this reason, payback period is often used as a measure of risk.  We generally see companies looking favorable on projects with a payback period of 1-3 years, with skepticism of projects over 3 years.

But consider two investments – one for $1M and another for $100K.  The first one has a 4 year payback (I get back, on average, $250K each year).  The second project has a 2 year payback (I get back, on average, $50K each year).  See the problem?

Although the payback period is easy to understand and communicate, it has serious limitations because it does not account for the time value of money.  A way around this is to use a measure for a Discounted Payback Period. Discounted payback period use a discounted cash flow technique to find the time period in which present value of future cash flows equals the initial cash outlay.  But once this metric is employed, the all simplicity of the payback period is lost.  If you are going to go to that much trouble, there are better metrics, in my opinion, that can be employed.

Another problem with payback period analysis is an assumption that project returns will continue even after the payback period is attained.  And payback period analysis does not specify any required comparison to other investments – or even to not making an investment at all.

Fortunately, there are other, alternative measures of a project’s return to consider.  Two of these are net present value (NPV) and internal rate of return (IRR). Although these metrics have problems and limitations of their own, they are much stronger economic analysis metrics.

• The 2011 deduction limit was raised from $250,000 to $500,000.  This is good on new and used equipment and includes software purchases like Warehouse Control Systems.
• The 2011 limit on equipment purchases to qualify for the deduction was raised from $800,000 to $2 million dollars.
• Here is the best part.  A “bonus” depreciation of 100% on new equipment purchases can be taken after the $500k deduction limit is reached.  This allows businesses that exceed the $2 million cap to write-off 100% of qualified assets IN THE FIRST YEAR.

For businesses that purchase equipment or software totaling $500,000 or less in 2011, the entire cost can be written-off on the 2011 tax return.  Instead of depreciating only, say 20%, the entire amount can be written-off in the first year, thereby reducing the tax burden of the company.  And businesses that exceed the $2 million in capital expenditure threshold can take a bonus depreciation of 100% on the amount that exceeds the limit.  Wow, that is a huge incentive to encourage businesses to buy equipment now.

Here is an example:

Suppose your company purchases a conveyor system totaling $1,000,000 in 2011.  The new law would allow you to write-off $500,000 on your 2011 taxes.  Additionally, you can deduct the remaining value ($500,000) using the bonus deprecation provision – thus writing-off the entire purchase in 2011.  If your corporate tax rate is 35%, then the tax burden is reduced by $350,000 (0.35 x $1,000,000).  So the new conveyor system essentially cost you only $650,000 ($1,000,000 – $350,000)!

Now, I am not a tax accountant, so you will want to check me out on this.  However, I think I am on base here.  But time is of the essence to take advantage of this tax law change as the new equipment has to be purchased by year-end.  With things heating up in the material handling industry and backlogs increasing – don’t get caught sleeping on this one!

I am going to share some of the main thoughts that came out of the audit I did for my client.  Hopefully, if you are involved in a similar operation, you might find a nugget or two as well.

 Labor Savings Opportunities 

As mentioned, valued added assembly is very labor intensive.  Operations need to be streamlined and run very lean if they are going to be effective.  Here are some ideas to consider to reduce the cost of value added assembly labor: 

Design, teach, and use good work flow.  Because the items being handled are always changing in a value added assembly area, it is imperative that the work orders are set-up so that the workers can be productive.  If you have a flexible assembly area, using nothing more than table space, the work orders should set-up to replenish from one side and assemble from the other side.  The layout would look something like this:

Although this should be intuitive, it is also important that the components be set-up so that the assembly flows from one end of the assembly table to the other.  Believe it or not, I have seen this not be the case.  Component product should also be placed in good relation to the assembly workers, to minimize their movements.  They shouldn’t have to walk around each other, reach across each other, etc.  The set-up must also be thought out so that the number of “touches” is minimized.  This requires some planning and experience. 

Bucket Brigade Assembly.  A concept that has taken hold in recent years is “bucket brigade” assembly.  This is a concept that was first developed at Georgia Tech with the goal of producing a self-leveling assembly line.  It is widely used today.  If you have ever been to a Subway resturant, it is the way that they assemble your sandwich (see www2.isye.gatech.edu/~jjb/bucket-brigades/case-studies/subway.html).  Although it is low-tech, it promotes the flow of the assembly build process.  More information can be found at: www.bucketbrigades.com. 

Install group leaders with accountability for production.  These people should be empowered to lead a small group that is fairly autonomous, that works together as a team, and that is accountable for their production.  The group leader should be a “working lead” that sets-up the work flow for the assembly orders that they are assigned.  They should set-up their team to maximize the productivity of each work order.  However, to get the most out of this, the group leader must be given some training on how to do this. 

Install a time keeping & labor tracking system.  Every employee and group should be measured against a goal time (and rate) for each work order done in the assembly area.  The time each person spends on each WO should be tracked.  As computerized time tracking system can be used to do this.  A simple one that works well for this type of environment is Time Clock Plus (www.timeclockplus.com).   

Consider paying a piece rate incentive.  With the tracking of piece rates by employee (or group), a piece rate incentive program can link the company’s goals (reducing the cost per unit) with the employees’ goals (ability to maximize their paycheck).  Based on the number of component pieces completed by the employee (or the employee’s group) and the number of hours logged by the employee for the week, the employee could get an incentive component of their compensation.  This should be designed not to increase the nominal wage of the mediocre employees, but rather provide a bonus to the exceptional performers. 

Make early decision on non-productive workers.  In most operations that have a regular valued added assembly area, the work force is mainly temporaries.  If it is determined that a temporary worker is not going to meet the productivity requirements of the operation, get them out of there early!  In most cases, a supervisor or group leader can tell if someone is going to be productive enough to keep within one hour – certainly by the first break.  Agreements with temporary agencies often have a one hour, no questions asked guarantee for this reason (if yours doesn’t, talk to your temp agency about getting this added).  An early decision on these people not only reduces the cost per unit, it also makes everyone else aware that low productivity is not tolerated. 

Everyone is interested in “green” these days.  We highlighted several environmental / energy-saving concepts related to distribution.  In the conveyance side, we discussed the emergence of 24V technology and its impact on power reduction. 

We were excited to present innovations in lighting, including new fluorescent and LED lights, advanced controls, and sky-lights.  We even discussed the impact of low-tech solutions like big fans and bailers.

Highlighted in the conveyance portion of the presentation was the use of ARB (Activated Roller Belt) technology.  ARB uses the best benefits of belt and roller conveyors in a combined format.  It can be used for several applications including aligning, switching, case turning, merging, and sorting.  It is really slick.

On the storage side, we discussed the surge of AS/RS (Automated Storage and Retrieval Systems) implementation in the U.S.   As more of these types of automated systems are being installed, integrators are finding even more uses for the technology.  Deep lane storage, imbedded picking modules, and product buffering are some of the applications that AS/RS is being employed.  We also showed the next generation of AS/RS, capable of even greater flexibility at high throughputs.

 On the order picking side, I got the chance to show the Pick-n-Go system.  As you know from previous posts, I am a big fan of this system to aid in “tour picking” operations.  We also showed the next generation of pick-to-light called “pick-to-display”, which replaces the colored lights with an OLED display.

We discussed the coming-of-age of robotics in distribution – mainly in end-of-line palletizing of same or similar SKU environments.  We also showed how robots are now being implemented in mixed case palletizing systems.  We also showed video of how robots are being utilized in each picking operations.

We threw the Kiva system into this part of the presentation.  Kiva is really a cross between AGV (automated guided vehicle), robotics, storage and a picking system.  It is a very interesting technology.

We discussed a lot of other technologies including warehouse control systems, human machine interfaces, labor management, and packaging systems – just to name a few.

We had great attendance at the presentation and I think it was well received.  The PowerPoint was recorded to video and as soon as it becomes available, I will post the link to this blog.

Okay, it is hard to explain, but the video below does a much better job demonstrating the technology.  If you find yourself as interested in this system as I was, send me an email (mlogan@bastiansolutions.com) and we can talk about it, as we are now Pick-n-Go integrators!

http://www.youtube.com/watch?v=Mn7HOIPjExk

Here are the options that we came up with to improve the quality, in the order of increasing effectiveness and cost.

1. Pallet count audit – After the system says that all the cartons for an order have been sorted, the operator (or an auditor) can do a count of all the cartons on the pallet and compare this to the expected quantity.  Granted, this is can be hard to do, esp. with various sized cartons.  Also, it will not help in the situation where two cartons were “swapped”, making the counts right, but the orders wrong.
2. Use a check-off sheet or peal-away tab – If the system is pre-assigning which cartons will go on which pallet, a list of all the carton id numbers for each pallet could be generated.  The operator would then check-off the carton id number as they place the carton on the pallet.  This really bogs down productivity.  Another slight variation of this is where the carton label has a peal-away tab, showing the carton id, which can be stuck on the pallet list.  If you really want to get clever, the list could then be OCR scanned into a system that can recognize if any of the carton ids don’t belong on the list.
3. Scan to the pallet – You can use RF scanning to improve quality.  There are two variations of this method.  The first is when the WMS or WCS pre-assigns cases to pallets.  In this method, the operator scans the barcoded carton ID number and then the barcoded pallet id (usually hanging from the ceiling, right above the pallet).  A variation of this method is when the WCS or WMS does not pre-assign the cases to pallets, but allows the operator to build a pallet on the fly.  Cartons are continually added to the order pallets until they are all accounted for or when the operator believes that the pallet is “full” and starts another pallet for the same order. In either case, the system makes sure that the scanned carton belongs on the pallet.  If it doesn’t, the RF unit beeps or makes some annoying sound.
4. Voice Direct Put – Instead of looking at the RF unit to determine what pallet the scanned case belongs on, a voice directed system could be used.  Operators wear head sets that “tell” them the pallet ID number to put the cases on.  When the operator scans the barcoded carton ID number, the system tells them the pallet number.  Once the case is put on the pallet, the operator confirms the put by saying a check digit or some other confirming code to the system.
5.  Put to light – Instead of looking at the RF unit to determine what pallet the scanned case belongs on, a put-to-light system could be used.  Light indicators are hung over the pallet spots and when the operator scans the barcoded carton ID number, the appropriate light illuminates.  If multiple operators are sorting to the same area, multiple colored lights can be used with a different color assigned to each operator.  To confirm that they put the cases on the correct pallet, the operator “extinguishes” the light by pressing the lighted button.
6. Build one pallet per sortation lane – A good way to remove the human error is to reduce the variability.  The system could be redesigned so that there is only one pallet at the end of the sortation lane – making a one-to-one sort.  If there are no changes made to the equipment side, this would require batch picking in significantly smaller waves – increasing operational costs.  Otherwise, either more sortation lanes are needed or some type of buffering and sequencing system is needed between picking and sortation – increasing capital costs.
7.  Robotic Palletizing – The best way to remove the human error is to remove the human.  Mixed case palletizing has come a long way.  Robots are better, end-of-arm tools are better, vision systems are better, and warehouse control systems are better – making this technologically feasible.  The capital costs are definitely higher than the other options, but the quality results are also much higher while improving the labor and ergonomic costs associated with palletizing operations. 

These are the options that I came up with.  I would love to hear from anyone that can think of other ideas!

I believe that inventory accuracy is the one area in a distribution center that can improve operational performance the most. And I am defining “inventory accuracy” as all information about inventory being accurate. This includes not only quantities, but accurate SKU numbers, barcodes, carton marking, lot codes, putaway rules, etc. – Basically any information tied to an item in the distribution center.

Think about all the problems that could be eliminated if your inventory were accurate. Receiving re-handling because of bad vendor information would be eliminated; product would be put away correctly from the receiving dock; items would always be available in the right locations to fulfill orders; product would never be left in locations that are supposed to be empty; inventory wouldn’t get “lost”; and orders could be shipped complete, without delays caused by shortages.

If this is the goal, then what are the steps necessary to get there? Well, here are some tips that I have come across that help to improve inventory accuracy:

• Proper Ownership. The first step in improving inventory accuracy is recognizing that it is related to operational performance (like the director in my example) and then taking ownership to make the required improvements (unlike the director in my example). Inventory accuracy is not the accounting manager’s or the inventory manager’s issue – it is the operational manager’s issue. And striving for inventory accuracy must be an integral part of the attitude of the organization. This attitude must start at the top.
• Cradle to Grave Processes. All process that effect inventory need to be clearly defined throughout the organization. These should be documented and the documentation should be as clear and comprehensive as possible. Key employees need to help develop these processes. All employees need to be trained and tested on them. Once employees are considered competent, management needs to “man-up” and hold people accountable. The process procedures need to be enforced.
• Good Gate-keeping. Receiving has to be the “narrow gate” where inventory accuracy starts. Product must be correctly identified and correctly received at the inbound docks. This requires both vendor and internal company participation. Employees on the receiving dock are critical positions and are the first level of inventory control. They need to be detail oriented and vigilantly look for informational discrepancies. To retain the position, a receiver should not make more than one error per quarter. If he/she does, he/she should be removed from the job. Ya, that’s how important it is.
• Receiving put-away verification. Even distribution centers with good warehouse management systems and good inventory processes do put-away verification. Everything can go right at receiving, and then the product gets put-away to the wrong location. Put-away verification is easy to do and just insures that the product was located to the right place. It saves the trickle-down effect of mistakes made at put-away.
• Zero Count Inventory. Most best-of-class distribution centers are cycle counting their inventory. No big new here. However, not all are doing zero counts. Basically, this requires that picking locations are random. When one goes “empty” (according to the WMS), the WMS assigns that SKU to another location. The zero count takes place by having some verify that the empty location is really empty. This validation not only uncovers quantity problems that would be buried if more inventory were just added to the original location, it also can be used to troubleshoot problems with the process or procedures.
• Daily Location Auditing. Just do it. I know that it is tough and requires you to put valuable headcount to the task. But I guarantee that your inventory your problems will begin to disappear. You can develop a report to do this or use your automation (RF handhelds, Pick-to-light, voice system), but the idea is to verify that the correct item is in the correct location. You don’t even have to do a count. Cycle through your warehouse in one to two weeks, auditing the inventory information (SKU, Lot number, etc.) in your locations. Then begin all over again. You will see benefits within a month.
• Hire an Anal Manager. Your inventory manager should be the strongest manager on your team. She/he should also be the most obsessively detailed to ensure consistent process application. This person needs to be technically competent, understanding your WMS and ERP systems, and how they treat inventory. She/he needs to be able to generate daily reports and know how to take action based on the information they provide.

Again, inventory accuracy is really an operational issue. Poor inventory accuracy is the root cause of many operational problems and needs to be given a high priority in the distribution center. The smart manager will recognize this and take ownership.

Virtual is the Real Way to Go

When the system was first implemented, we ran it for a period of time in a “hidden” mode.  The system only told the employees the target time for each task, but did not give them any feedback on their performance.  We did this in order to work out any bugs that might be in the standards.  We collected the target and actual information for analysis only.  When we were confident of the standards, we turned the feedback system on, allowing employees to see the goal times and their performance against the goals.

An interesting thing happened.  Overnight, we got a sustained 20% improvement in productivity.  Across the board, almost every employee improved their performance against goal.  The only thing that changed was completing the feedback loop.

I know that many if not most distribution organizations post some level of productivity information.  It is good management to keep people informed of goals and how the organization is doing against those goals.  But, unless an employee can understand how he/she individually contributes to those goals, the information may be interesting, but of little use to them.  The more discrete the information feedback loop is, the more they can affect it.