The development of uniform data standards for data recorded in electronic vending machines has created a wealth of opportunities for managing the data more efficiently.
In time, as more operators become familiar with these benefits and begin using them, they will be able to maximize the return on their service labor, menu their machines to meet location needs more accurately, utilize cashless solutions more economically, and be privy to a host of other benefits associated with line-item tracking.
Last month, I explored the differences between the three technologies that have emerged — multi-drop bus (MDB), data exchange (DEX) and data transfer standards (DTS) — and how they interact in a vending machine. Much of the article addressed the distinctions between DEX and MDB. This month, I will focus more on the collection methods and benefits afforded by DEX data collection, many of which are only in the experimental stage.
DEX data collection
There are three levels of DEX data collection: machine, local and remote.
Machine data capture involves either plugging a handheld device in a machine's DEX port or using an optical alternative, such as infrared technology.
Many DEX ports enable the vending machine controller (VMC) to detect the insertion of the handheld device plug so that it serves as a signal to start the DEX data gathering process.
Local machine polling incorporates a handheld device (or pocket probe) designed to connect to a vending machine's DEX port or to communicate through an IR port. Once the connection is established, the device is used to extract (upload) transactional data from the machine to the handheld device.
A typical DEX data upload takes approximately five seconds. Field collected data can be transferred from the handheld device to a central office computer for processing, analysis and report generation.
Wireless technology makes it possible to collect DEX data from the comfort of a route truck located within location proximity. Some operators incorporate truck-based replenishment ticket printing and/or enable a handheld device to capture machine data to provide the validity of polled curbside data.
Like other remote options, curbside polling eliminates at least one trip back to the truck when filling machines.
In order to extend the capabilities of curbside polling, some operators are trying to apply warehouse fill orders or dynamic scheduling without intensive forecasting technology. Curbside polling is normally considered best for daily, unpredictable inventory machines. Curbside polling does not usually incur line charges or monthly fees within the network, and is the least expensive polling option.
Dial-up polling involves use of a modem and telephone line. Once a valid connection is established, DEX data can be transported to a remote office or warehouse location for evaluation over an Internet or virtual private network (VPN) connection.
Dial-up polling may require a dedicated phone line connection for each vending machine, or simply one line connected to a master DEX data consolidation device. Phone line availability, line charges, service fees, location permission, and the like may present barriers to successful connectivity.
Dial-up polling enables a machine to be remotely monitored with respect to cash, inventory, machine alerts and malfunctions.
Usually a single machine with a cellular modem, located in a cluster of machines, can serve as a master unit for data transference and machine monitoring. The master (or host) unit gathers DEX-data-linked (slave) units within an approximate 1,000-yard radius. Readings can be performed hourly, or at user-defined intervals.
Cellular modem applications tend to be more cost-
effective in locations where remote polling is problematic or placing a land line is impractical. A cellular modem
typically involves recurring monthly service fees and related expenses.
Wide area network polling
In a wide area network configuration, one machine is designated as the master unit and functions as a data transmitter. The master unit is capable of transmitting and receiving data from three to 10 miles, depending on topography. The master unit is designed to constantly search for slaves and other masters so an effective DEX data relay can be established.
If there is no local Internet connection, the master unit will pass its accumulated readings to the next master between it and the vending operator's warehouse. Depending on design, the warehouse may require an antenna.
Similar to other forms of polling, wireless polling enables remote access to DEX data via a network. Wireless polling, however, relies upon network connectivity to establish the proper linkage.
Many wireless methods of DEX transfer include file compression or filtering to reduce file size (i.e., the number of bytes transmitted) as a cost containment strategy. The advancement of wireless technology has emerged as an attractive alternative.
Wireless applications possess tremendous potential for the vending industry, an industry that desires mobility, flexibility and reliability in enterprise-wide operations.
Vending practitioners dissatisfied with the constraints and complexities of hard wiring are migrating to the convenience of design portability and user mobility that wireless technology solutions provide.
Data collection models
Several factors contribute to the viability of remote data collection and monitoring:
- Cost containment. Hardware and communication expenses (e.g., landline modem, wireless cellular modem and Ethernet cabling) have declined
as technology becomes more standardized and widespread.
- Interconnecting machines. Configuring a group of vending machines to form a bank, or block, capable of sharing a single point of connectivity minimizes expenditures.
- Integrated application software. Industry software suppliers have developed integrated remote monitoring software or have partnered with remote monitoring companies to form an integrated package.
- Cashless vending. Since cashless vending settlement involves a communication link to authorize payment processing, the same technology can be used to transport DEX data to a vending management software system, at little or no additional cost.
Remote DEX-data transmission typically is conducted at predetermined intervals (e.g., nighttime is usually preferred as polling may render the vending machine inoperable for 10 to 40 seconds during the process) and is dependent on sensitive connectivity prone to communication outages and damage or displacement by weather or vandalism.
Since most drivers use handhelds for inventory, product planogram changes and collecting additional data (like refunds and test vends) during the day, a majority of operators elect to continue using handheld devices despite the fact newer machines can be built with remote connectivity options.
One of the principal reasons vending management software utilizes DEX data is to control cash shortages and product shrinkage.
The table on page 36 summarizes a sample of available DEX data collection methods.
Activity/error reporting is an important DEX function because a large portion of the wireless vending benefit relies heavily on the timely reporting of activity and errors.
It is estimated there are many possible pieces of information that can be exchanged between a vending machine and a polling device; however, a set of eight are usually selected for transmission. The key variables are listed above.
A DEX file contains a set of vending machine transactions and event data. Stored data includes units sold by column, transactions by cash and credit, and various error codes measured in interval (since last reading) and cumulative metrics.
By comparing two consecutive DEX file readings, an operator can evaluate changes in cumulative totals.
In its raw form, DEX data is not immediately useful, and therefore, operators interested in developing meaningful DEX information typically expand the analysis to include:
- Transmission of the DEX file to the operator.
- This can be done using a handheld device at the machine, or via local or remote networking.
- Application of vending management software to interpret DEX data. Vending management software relies on formulations and report formats to translate raw data into useble information.
- Training and operational support to enable implementation of product sales by correlation to a planogram or product mapping.
DEX file variances
The structure and content of DEX files can be different by type of machine (e.g., coffee brewer versus cold beverage) or by manufacturer and model. Such diversity must be recognized and addressed.
Given this potential lack of uniformity, the cost and time required to implement a full-featured vending management system, operators may opt instead to focus on the transactional data generated through a MDB payment device.
MDB data is stored as a complete electronic transaction record that may include: date, time, number of transactions, vend column and price. This record can be retrieved remotely on a daily basis when connected to a communications network.
A communications network designed to collect MDB data is available as an additional benefit of cashless vending technology. Since cashless vending relies on MDB data to process credit card transactions, it can also be used as a stand-alone audit device (in place of DEX).
Providers of remote MDB reporting technology can provide transactional records via a Web-based reporting system, thereby eliminating the need for a separate route management system.
The information provided, while somewhat less detailed than DEX data, provides sufficient information for management to conduct route settlement with full accountability for all transactions, both cash and credit.
MDB-based reports can provide sales by machine, location, route or total operation, and by settlement type, time-of-day and multiple machine reporting. Reports can be used to help identify out-of-change conditions, low product inventory and out-of-service conditions.
Item level tracking: "V-engineering"
Tracking item level data for each machine allows for more effective replenishment, product rotation and improved profitability. This is the concept of "v-engineering." Since neither DEX nor MDB data is capable of identifying products sold without correlation to a planogram or product map, item level sales analysis can be tedious.
Similar to the popular menu engineering model used in commercial foodservice, v-engineering relies on a database of historical item sales data to identify best sellers, compute comparative profit margins, and formulate informed decisions relative to future product offerings.
Evaluative metrics can be applied by individual machine, location, account and route. Effective metrics include:
• Percent depleted — How much inventory remained in the machine at the time of servicing.
• Number of sold-outs — How many sold-out columns (and more importantly, products) were in the machine at the time of service.
• Percent filled — How filled the machine was following servicing.
• Average daily sales —The velocity of product movement.
Vending management software is capable of aggregating these metrics to reveal trends and opportunities.
V-engineering involves replacement of slow moving products with faster moving items to enhance machine contribution margin while providing for more efficient machine servicing for enhanced customer satisfaction. In the future, the application of v-engineering concepts should lead to increased sales, improved contribution margins and extended customer services.