Selling More To Upstream and Downstream Supply Chain Partners

It is common to see companies supplying to both upstream and downstream supply chain partners.

For your bigger customers, it is worth mapping out what kinds of products upstream and downstream partners along a supply chain are taking from you, especially for those customers are serving end customer group.

The very least you can get out of this mapping is you will know what partners nearer to end customers are cooking themselves and not taking from upstream suppliers, and who are buying products just to build on or customize products developed by upstream suppliers.

For supply chain partners developing end products by themselves and not taking from suppliers, you can then run it by your bill of materials (BOM) receipe database to suggest what new products that they can develop. In that way, you can broaden the product range ordered from your company.

For supply chain partners customizing/finishing up on earlier partners' products, you can again run your BOM receipe database to look at what products can be customized nearer to end customers to sell to upstream and downstream supply chain partners. Especially BOM receipes that involves minimal processing and are fast.

Both this strategies help to broaden the SKUs and increase its volume for your customers. It also deepens the relationship between you and your customers along the supply chain.

To enable this, one will have to build up their BOM receipes and link it to their SKUs. It can be done industry knowledge and by trawling the Internet.

Practical Re-order Level Setting

When lead-time and demand variability is variable, it is hard to define a re-order point, even when the cycle service level is clear.

As such, a practical way to set reorder points is to tap on the experience of the warehouse staff to identify out of stock (OOS) SKUs, and to go through the demand variability, lead time, cycle service level to set new reorder points and monitor OOS. The staff may tell you that they previously order the minimum order quantity, or there are highly variable demand or lead times? With experience, you can also set higher safety stock, and hence a higher reorder point for seasonal products.

How do you free up space for higher reorder points? Well, the same experienced staff will be able to ye you what SKUs could be over stocked. You can then calculate a lower reorder point to free up space.

Transhipment Supply Chain Bar-Coding For Track & Trace and Supply Chain Performance

Knowing how we could incorporate possible transhipment points in your suppy chain network calculations to speed up, pool risk to lower inventory cost and postponing customization isn't enough.

There growing need for track and track for sub-components of a finished project for patient/food/consumer safety. As such, there is an increasing need to add in more information on top of current product bar codes.

Besides tracking and tracing, the additional plant and production/expiry date/time information of sub-components can also be used to measure supply chain velocity, quality and costs.

So the question is, how do we add in more information from say, a transshipment hub like Singapore?

How about using EAN 128, or GS1 128 bar codes? It has SSCC, GTIN, Country Of Origin, Best Before etc... even optional fields you can define. It's open standards ensures interoperability worried about the length of bar codes to scan? You can use a QR code to store much more information in a smaller space. QR code specification is free and there are tonnes of free QR code generators around, including a Google API for QR. Multiple products on 1 QR code? No problem, we just need to configure the reader to consider it as separate product if 00,01 and 02 application identifiers are read, as these application identifiers will have unique company prefix.

You can use my web application at www.transhipmentflow.appspot.com to decode your EAN 128 or GS1 128 bar codes into its application identifiers and know how much more information you can add if you convert to using QR codes.

Consumer Analytics To Complement Marketing and Sales Strategies

Regression to identify significant drivers of sales, forming hypothesis with significant factors, then using naive bayesian classifier to test hypothesis is quick and reasonably good approach. Especially so for hub and spoke operations in Singapore where forecasting regional needs are challenging. It can also be used to compare performance between stores and where new stores should be opened.

Naive Bayesian classifier has been found to perform well against decision trees and neural network classifiers. A good example of naive Bayesian classifier, including Laplacian Correction, can be found at http://cis.poly.edu/~mleung/FRE7851/f07/naiveBayesianClassifier.pdf

Marketing budgets may be allocated base on rate of change of probabilities with time, so the above approach helps companies to obtain quantitative comparisons weekly or every 2 weeks. Even for probability in the first pass, the difference in null and alternate hypothesis already helps to clarify the strength of the null or alternate hypothesis.

Transshipment Flow Analytics Approach

Adding in a transshipment point to transportation problem is a common approach. Then the transshipment point will allow equal capacity volumes 'in' and 'out' of the transshipment point from different sources to different demand (sink) points.

Base on profit margin (marginal contribution) approach, the demand at sink points can also be modified in the model.

Of course, safety stock will be reduced with reduced lead time and propagation of variability. The cost of pooling variability in a transshipment point like Singapore can be compared to cost of holding more stocks regionally.

Slot & Pick By Ascending Sizes Along The Height Of Rackings

For retail products like apparel and bags, it makes sense to slot it by common sizes, or even commonly sold colors horizontally. For example, in the case of shoes, the smallest sizes can be at the bottom-most racks, common sizes at the middle (easy to pick) racks, and larger sizes at the upper racks.

In this way, put-away and picking is not difficult as sizes go up with racks. This also allows more SKUs to be placed along easy to pick level of shelvings/rackings.

Pick List with Bar Codes to facilitate picking

Many companies still generate paper pick lists. It is cheap - there is no need to lug a handheld around or wear a headset.

If you have the product bar code on the pick list, it is probably faster to scan product bar code on pick list than on actual products. As location is usually on the pick list itself, picking errors is unlikely anyway.

It is quite easy to generate the equivalent bar code on the pick list, especially of the pick list can be saved as an excel sheet. You can customize the bar codes you need too.

Online Product Sales Prediction

Good forecasts reduces excess inventory that would require markdowns and disposal. There are also cost of space and time/system used to monitor inventory. The savings are significant when you have a huge number of SKUs and/or carry lots of inventory.

One approach I had developed for an global data science competition is as below. It gives pretty good results. It is easily understandable (keep it simple approach), one can compare the accuracy using root mean square error (RMSE) with your current forecast, and most interesting, evolve your own algorithm to give weightage or overrides base on how sales evolves. I achieve an RMSE of about 1.5 :)

Step 1 - First review all the SKUs its date, quantitative or category variables to decide which variables(columns) to ignore, and which outcome(rows) to delete. With a huge number of SKUs, you certainly can't focus on all of them, so you can apply the Pareto rule. If you only have one or a few SKUs, you can skip this step.

Step 2 - Apply regression on each monthly/weekly/daily sales for each SKU as outcomes using the remaining variables as inputs. Remove non-significant factors and run the regression again to get the adjusted coefficients. If you have more than 16 variables, use minitab, SPSS, SAS or r.

Step 3 - Apply regression equation to only the significant factors to come out with outcomes on monthly/weekly/daily basis. Look at your significant factors with time. Does it make sense and did you get new insights in how your sales are affected by different variables.

Step 4 - Use RMSE to compare this new forecast with your older forecast. If the forecast is not better, adjust using your knowledge of the market using weightage, omitting factors, averages or comparison to multiple of averages.

Step 5 - Add in marginal contribution and time series analysis if you want to factor in profit margins and time trends.

Step 6 - Implement changes proposed significant factors identified from the analysis done the the above steps for continuous, self-adjusting capability.

Step 7 - Synchronize sales with operations and capacity planning.

Write to me if you need more help! :)

Facilitate Picking Using Invoices with ERP systems

Rather than using pick lists, some companies pick products using one ply of invoices. For a distribution operations dealing with a huge variety of SKUs, it is likely that SKU number would be classified according to product groups like (dry goods, chilled or frozen products). To facilitate picking, it may be better for the SKU number on invoices to be shorted alpha-numerically.

For example, in Microsoft GP, sorting by SKU number on invoice is
MS Dynamic GP> Tool > Customise > Report Writer > Reports > Invoice History & New Invoice > Sort > Choose Item number column > Descending